CFTC Project
Kevin Williams
2024-10-06
Project Development for plotting and analyzing CFTC data.
Premise: Commitments of Traders Futures Data may give insight into future price action for some Fx Pairs and commodites. This information may possibly be packaged in a way that could be sold for a subscription.
Background: Steven Jon Kaplan (SVK), who runs the website and service
at https://www.truecontarian.com, frequently uses extremes
in commitments of traders for commercial and non-commerical traders to
make trades. Recent example as of 15 Sep 2024: Steven is long palladium
do to extreme long interest by commercial traders.
This project will attempt to collect data on commitments of traders and
display in a manner usable for making investment decisions.
Ideas:
29 SEP 2024:
-Show commercial and non-commercial traders by long and short
-Show net long or short for traders
-Show a ratio of long vs. short by trader type
-Find historical extremes; perhaps by last year, last xx years, all
time
-Compare to price
-Make maximum gain or loss over the next xx months; 3, 6, 12, 18?
Example
This screenshot below is from the website: http://cotpricecharts.com SVK subscribed to this and
would occasionally share some information from the website and would
sometimes make trades when commitments of traders seemed to hit extremes
for some assets.
The website is no longer active. The person who maintained the webiste
and updated it passed away and no one took his place to maintain it.
https://www.appsilon.com/post/r-markdown-tips#images
https://www.earthdatascience.org/courses/earth-analytics/document-your-science/add-images-to-rmarkdown-report/
Old example from http://cotpricecharts.com by Software North
To do: get a picture of a current website.
xxxxxxx
www.cftc.gov has this kind of data. Looks like this…
To do: show a screenshot of current data from cftc.gov
Goal: create something that might look like this but also shows max gains and losses. Perhaps out to 6 months and 12 months???
Step 1: look around www.cftc.gov
Go Commitments of Traders here: https://www.cftc.gov/MarketReports/CommitmentsofTraders/index.htm
Credit to Steven Jon Kaplan of True Contrarian at https://truecontrarian-sjk.blogspot.com
Step 2: Figure out webscraping.
Here are websites for help with webscraping:
https://r4ds.hadley.nz/webscraping
https://preettheman.medium.com/best-web-scraping-packages-for-r-2787979b747
https://scrapingant.com/blog/r-web-scraping
https://steviep42.github.io/webscraping/book/
https://statsandr.com/blog/web-scraping-in-r/
https://www.zenrows.com/blog/web-scraping-r
https://www.bigbookofr.com/chapters/getting%20cleaning%20and%20wrangling%20data
https://oxylabs.io/blog/web-scraping-r
https://brightdata.com/blog/how-tos/web-scraping-with-r
15 Sep 2024: Might not need web-scraping as there is a new file each week.
Step 2 complete/not required
Get something working. Because SVK is current active trading palladium, we will start with that commodity.
Download data. Historical and recent data that can be downloaded in Excel format can be found on this link: https://www.cftc.gov/MarketReports/CommitmentsofTraders/HistoricalCompressed/index.htm
We download all of these files except the one “Starting July 20,
2010”.
The 2006-2016 file
actually downloads a folder containing three files with different years:
1986-2006, 2007-2014, and 2015-2016. We save all of these files to our R
working directory and rename them by year.
Screenshot of the sizes of files from
cftc.gov
As one can see, they files are a bit large.
Picking one at random (2015-2016), let’s get a plot of something.
But first, what is the Commitment of Traders data? Here is what recent data from cftc.gov webiste looks like for palladium:
Screenshot of Palladium Commintments of Traders
from 17 Sep 2024
Generally:
- “Non-Commercial” are entities like hedge funds and those who manage
other people’s money
- “Commercial” are those that use palladium as part of their business
and banks
- “Nonreportable” are individual traders
Let’s look at Commercial
https://www.tidyverse.org/packages/
library(tidyverse) # to use tidy data ## ── Attaching core tidyverse packages ──────────────────────── tidyverse 2.0.0 ──
## ✔ dplyr 1.1.4 ✔ readr 2.1.5
## ✔ forcats 1.0.0 ✔ stringr 1.5.1
## ✔ ggplot2 3.5.1 ✔ tibble 3.2.1
## ✔ lubridate 1.9.3 ✔ tidyr 1.3.1
## ✔ purrr 1.0.2
## ── Conflicts ────────────────────────────────────────── tidyverse_conflicts() ──
## ✖ dplyr::filter() masks stats::filter()
## ✖ dplyr::lag() masks stats::lag()
## ℹ Use the conflicted package (<http://conflicted.r-lib.org/>) to force all conflicts to become errorslibrary(readxl) # to read in Excel format data
# Read Excel file
fullFile_FUT15_16 <- read_xls("FUT15_16.xls")
fullFile_FUT15_16## # A tibble: 24,044 × 126
## Market_and_Exchange_Names As_of_Date_In_Form_Y…¹ Report_Date_as_MM_DD…²
## <chr> <chr> <dttm>
## 1 WHEAT-SRW - CHICAGO BOARD OF T… 161227 2016-12-27 00:00:00
## 2 WHEAT-SRW - CHICAGO BOARD OF T… 161220 2016-12-20 00:00:00
## 3 WHEAT-SRW - CHICAGO BOARD OF T… 161213 2016-12-13 00:00:00
## 4 WHEAT-SRW - CHICAGO BOARD OF T… 161206 2016-12-06 00:00:00
## 5 WHEAT-SRW - CHICAGO BOARD OF T… 161129 2016-11-29 00:00:00
## 6 WHEAT-SRW - CHICAGO BOARD OF T… 161122 2016-11-22 00:00:00
## 7 WHEAT-SRW - CHICAGO BOARD OF T… 161115 2016-11-15 00:00:00
## 8 WHEAT-SRW - CHICAGO BOARD OF T… 161108 2016-11-08 00:00:00
## 9 WHEAT-SRW - CHICAGO BOARD OF T… 161101 2016-11-01 00:00:00
## 10 WHEAT-SRW - CHICAGO BOARD OF T… 161025 2016-10-25 00:00:00
## # ℹ 24,034 more rows
## # ℹ abbreviated names: ¹As_of_Date_In_Form_YYMMDD, ²Report_Date_as_MM_DD_YYYY
## # ℹ 123 more variables: CFTC_Contract_Market_Code <chr>,
## # CFTC_Market_Code <chr>, CFTC_Region_Code <chr>, CFTC_Commodity_Code <chr>,
## # Open_Interest_All <dbl>, NonComm_Positions_Long_All <dbl>,
## # NonComm_Positions_Short_All <dbl>, NonComm_Postions_Spread_All <dbl>,
## # Comm_Positions_Long_All <dbl>, Comm_Positions_Short_All <dbl>, …This has all data. 24,044 rows and 126 columns. And in tibble data
format.
Limit to palladium:
# Just palladium
palladium_FUT15_16 <- fullFile_FUT15_16 %>% filter(str_detect( Market_and_Exchange_Names, "PALL"))
palladium_FUT15_16## # A tibble: 104 × 126
## Market_and_Exchange_Names As_of_Date_In_Form_Y…¹ Report_Date_as_MM_DD…²
## <chr> <chr> <dttm>
## 1 PALLADIUM - NEW YORK MERCANTIL… 161227 2016-12-27 00:00:00
## 2 PALLADIUM - NEW YORK MERCANTIL… 161220 2016-12-20 00:00:00
## 3 PALLADIUM - NEW YORK MERCANTIL… 161213 2016-12-13 00:00:00
## 4 PALLADIUM - NEW YORK MERCANTIL… 161206 2016-12-06 00:00:00
## 5 PALLADIUM - NEW YORK MERCANTIL… 161129 2016-11-29 00:00:00
## 6 PALLADIUM - NEW YORK MERCANTIL… 161122 2016-11-22 00:00:00
## 7 PALLADIUM - NEW YORK MERCANTIL… 161115 2016-11-15 00:00:00
## 8 PALLADIUM - NEW YORK MERCANTIL… 161108 2016-11-08 00:00:00
## 9 PALLADIUM - NEW YORK MERCANTIL… 161101 2016-11-01 00:00:00
## 10 PALLADIUM - NEW YORK MERCANTIL… 161025 2016-10-25 00:00:00
## # ℹ 94 more rows
## # ℹ abbreviated names: ¹As_of_Date_In_Form_YYMMDD, ²Report_Date_as_MM_DD_YYYY
## # ℹ 123 more variables: CFTC_Contract_Market_Code <chr>,
## # CFTC_Market_Code <chr>, CFTC_Region_Code <chr>, CFTC_Commodity_Code <chr>,
## # Open_Interest_All <dbl>, NonComm_Positions_Long_All <dbl>,
## # NonComm_Positions_Short_All <dbl>, NonComm_Postions_Spread_All <dbl>,
## # Comm_Positions_Long_All <dbl>, Comm_Positions_Short_All <dbl>, …A quick first plot of Commercial Long:
https://ggplot2-book.org/layers
# Plot long commercial positions vs date
library(ggplot2)
comm_long <- ggplot(palladium_FUT15_16, aes(x=Report_Date_as_MM_DD_YYYY,y=Comm_Positions_Long_All)) +
geom_line()
plot(comm_long)
Add Commercial Short to plot:
# Next, add commercial short positions also on same plot
comm_long_short <- ggplot(palladium_FUT15_16) +
geom_line(aes(x=Report_Date_as_MM_DD_YYYY,y=Comm_Positions_Long_All)) +
geom_line(aes(x=Report_Date_as_MM_DD_YYYY,y=Comm_Positions_Short_All))
plot(comm_long_short)
Quick eyeball analysis for 2015-2016: Commercial Short positions are
ALWAYS larger than Commercial Long positions.
Need to clean up: add key for two lines, correct y-axis label, clean up
x-axis label.
Note: adding a legend for multiple lines in ggplot2 requires adding aes
info as discussed here: https://stackoverflow.com/questions/40833809/add-legend-to-geom-line-graph-in-r
and here: https://www.geeksforgeeks.org/add-legend-for-multiple-lines-in-r-using-ggplot2/
# Make a plot with palladium data
comm_long_short <- ggplot(palladium_FUT15_16) +
# line for Commercial Long
geom_line(aes(x=Report_Date_as_MM_DD_YYYY,y=Comm_Positions_Long_All, colour = "Commercial Long")) +
# line for Commercial Short
geom_line(aes(x=Report_Date_as_MM_DD_YYYY,y=Comm_Positions_Short_All, colour = "Commercial Short")) +
# add label
labs(
y="Commercial Interest",
x="Date"
)
# show plot.
plot(comm_long_short)
That works.
Clean up some more:
# Make a plot with palladium data
comm_long_short <- ggplot(palladium_FUT15_16) +
# line for Commercial Long
geom_line(aes(x=Report_Date_as_MM_DD_YYYY,y=Comm_Positions_Long_All, colour = "Long"), size = 1) +
# line for Commercial Short
geom_line(aes(x=Report_Date_as_MM_DD_YYYY,y=Comm_Positions_Short_All, colour = "Short"), size = 1) +
# better colors and legend
scale_color_manual(name = "Commercial Interest", values = c("Short" = "red", "Long" = "darkgreen")) +
# add label
labs(
y="Commercial Interest",
x="Date",
title = "Commercial Futures Interest for Palladium As Reported by cftc.gov"
)## Warning: Using `size` aesthetic for lines was deprecated in ggplot2 3.4.0.
## ℹ Please use `linewidth` instead.
## This warning is displayed once every 8 hours.
## Call `lifecycle::last_lifecycle_warnings()` to see where this warning was
## generated.# show plot.
plot(comm_long_short)
That looks OK for now.
Lets look at a ratio of Commercial Long to Commercial Short
interest:
Need to use mutate:
https://cran.r-project.org/web/packages/dplyr/vignettes/rowwise.html
https://dplyr.tidyverse.org/articles/rowwise.html
# Add a ratio of long to short
palladium_FUT15_16 <- palladium_FUT15_16 %>% mutate(long_short_ratio = Comm_Positions_Long_All/Comm_Positions_Short_All)
comm_long_short_ratio <- ggplot(palladium_FUT15_16) +
geom_line(aes(x=Report_Date_as_MM_DD_YYYY,y=long_short_ratio))
plot(comm_long_short_ratio)
And also Commercial Short to Commercial Long:
# Add a ratio of long to short and short to long
palladium_FUT15_16 <- palladium_FUT15_16 %>% mutate(long_short_ratio = Comm_Positions_Long_All/Comm_Positions_Short_All) %>% mutate(short_long_ratio = Comm_Positions_Short_All/Comm_Positions_Long_All)
comm_long_short_ratio <- ggplot(palladium_FUT15_16) +
geom_line(aes(x=Report_Date_as_MM_DD_YYYY,y=long_short_ratio))
plot(comm_long_short_ratio)
comm_short_long_ratio <- ggplot(palladium_FUT15_16) +
geom_line(aes(x=Report_Date_as_MM_DD_YYYY,y=short_long_ratio))
plot(comm_short_long_ratio)
Looks OK.
Let’s clean up some:
comm_long_short_ratio <- ggplot(palladium_FUT15_16) +
geom_line(aes(x=Report_Date_as_MM_DD_YYYY,y=long_short_ratio)) +
labs(
y="Ratio",
x="Date",
title="Ratio of Commercial Long Interest to Commercial Short Interest - Palladium Futures"
)
plot(comm_long_short_ratio)
comm_short_long_ratio <- ggplot(palladium_FUT15_16) +
geom_line(aes(x=Report_Date_as_MM_DD_YYYY,y=short_long_ratio)) +
labs(
y="Ratio",
x="Date",
title="Ratio of Commercial Short Interest to Commercial Long Interest - Palladium Futures"
)
plot(comm_short_long_ratio)
Let’s try to plot all together:
Note: par will plot multiple plots together in the same image but is
base R and does not work with ggplot2. Need to try package cowplot or
patchwork.
https://cran.r-project.org/web/packages/cowplot/vignettes/introduction.html
https://patchwork.data-imaginist.com
# par(mfrow = c(2,2)) only works in base R
# need another package to link plots together: patchwork
# other patchwork info:
# https://cran.r-project.org/web/packages/patchwork/vignettes/patchwork.html
# https://patchwork.data-imaginist.com
# https://mattherman.info/blog/ppt-patchwork/
library(patchwork)
comm_long_short / comm_long_short_ratio
All three together:
# par(mfrow = c(2,2)) only works in base R
# need another package to link plots together: patchwork
# other patchwork info:
# https://cran.r-project.org/web/packages/patchwork/vignettes/patchwork.html
# https://patchwork.data-imaginist.com
# https://mattherman.info/blog/ppt-patchwork/
comm_long_short_ratio / comm_short_long_ratio / comm_long_short
Change size of ratio charts:
# par(mfrow = c(2,2)) only works in base R
# need another package to link plots together: patchwork
# other patchwork info:
# https://cran.r-project.org/web/packages/patchwork/vignettes/patchwork.html
# https://patchwork.data-imaginist.com
# https://mattherman.info/blog/ppt-patchwork/
comm_long_short_ratio / comm_short_long_ratio / comm_long_short + plot_layout(heights = c(1,1,5))
Try to put a ratio on the same chart as the Long and Short
interest.
This is actually very hard to do in ggplot2. Apparently, this is done on
purpose as the makers of ggpolot2 don’t approve of plots with two axes.
Seriously.
https://stackoverflow.com/questions/3099219/ggplot-with-2-y-axes-on-each-side-and-different-scales
# This y limit for long and short interest is static. Need to make dynamic.
ylim.prim <- c(0,30000) #max of comm positions
# This y limit for ratio is static. Need to make dynamic.
ylim.sec <- c(0,1) #ratio
b <- diff(ylim.prim)/diff(ylim.sec)
a <- ylim.prim[1] - b*ylim.sec[1]
comm_long_short_ratio_3 <- ggplot(palladium_FUT15_16, aes(x=Report_Date_as_MM_DD_YYYY,y=Comm_Positions_Long_All)) +
geom_line() +
geom_line(aes(x=Report_Date_as_MM_DD_YYYY,y=Comm_Positions_Short_All)) +
geom_line(aes(y = a + long_short_ratio*b), color = "red") +
scale_y_continuous("Comm Positions", sec.axis = sec_axis(~ (. - a)/b, name = "Ratio")) +
labs(
x="Date",
title = "Palladium Commercial Long & Short Interest and Ratio"
)
plot(comm_long_short_ratio_3)
This works. But let’s try an interactive plot with plotly for R. https://plotly.com/r/
library(plotly)##
## Attaching package: 'plotly'## The following object is masked from 'package:ggplot2':
##
## last_plot## The following object is masked from 'package:stats':
##
## filter## The following object is masked from 'package:graphics':
##
## layoutggplotly(comm_long_short)And with ratio?
library(plotly)
ggplotly(comm_long_short_ratio_3)That works, but will need some cleanup.
However, let’s try to stack the plots as before but make them
interactive as above and all linked.
https://stackoverflow.com/questions/58103196/how-to-facet-a-plot-ly-chart
fig1 <- ggplotly(comm_long_short)
fig2 <- ggplotly(comm_long_short_ratio)
figs_1and2 <- subplot(fig1, fig2)
figs_1and2Now stacked:
https://plotly.com/r/subplots/
https://stackoverflow.com/questions/32955163/multiple-ggplot2-plots-with-plotly?rq=4
fig1 <- ggplotly(comm_long_short)
fig2 <- ggplotly(comm_long_short_ratio)
figs_1and2 <- subplot(fig1, fig2, nrows = 2)
figs_1and2Adding a “unified hover”:
https://stackoverflow.com/questions/66767067/how-to-change-order-of-r-plotly-hover-labels-with-x-unified-mode
https://stackoverflow.com/questions/69903147/how-to-link-traces-between-subplots-in-r-plotly-with-shared-y-axis-so-that-hover
https://stackoverflow.com/questions/54561931/r-plotly-link-subplots-so-that-multiple-tooltips-shown-on-hover
https://plotly.com/r/hover-text-and-formatting/
https://plotly.com/r/reference/layout/#layout-hovermode
https://community.plotly.com/t/announcing-plotly-r-4-9-2-9000-shape-drawing-unified-hover-and-rangebreaks/40086
fig1 <- ggplotly(comm_long_short)
fig2 <- ggplotly(comm_long_short_ratio)
figs_1and2 <- subplot(fig1, fig2, nrows = 2) %>% layout(hovermode = "x unified")
figs_1and2fig1 <- ggplotly(comm_long_short)
fig2 <- ggplotly(comm_long_short_ratio)
figs_1and2 <- subplot(fig1, fig2, nrows = 2, shareX = TRUE) %>% layout(hovermode = "x unified", hoversubplots = "axis")
figs_1and2## Warning: 'layout' objects don't have these attributes: 'hoversubplots'
## Valid attributes include:
## '_deprecated', 'activeshape', 'annotations', 'autosize', 'autotypenumbers', 'calendar', 'clickmode', 'coloraxis', 'colorscale', 'colorway', 'computed', 'datarevision', 'dragmode', 'editrevision', 'editType', 'font', 'geo', 'grid', 'height', 'hidesources', 'hoverdistance', 'hoverlabel', 'hovermode', 'images', 'legend', 'mapbox', 'margin', 'meta', 'metasrc', 'modebar', 'newshape', 'paper_bgcolor', 'plot_bgcolor', 'polar', 'scene', 'selectdirection', 'selectionrevision', 'separators', 'shapes', 'showlegend', 'sliders', 'smith', 'spikedistance', 'template', 'ternary', 'title', 'transition', 'uirevision', 'uniformtext', 'updatemenus', 'width', 'xaxis', 'yaxis', 'barmode', 'bargap', 'mapType'The second plot seems to work as best as able to make it.
Add the second ratio for comparison:
fig1 <- ggplotly(comm_long_short)
fig2 <- ggplotly(comm_long_short_ratio)
fig3 <- ggplotly(comm_short_long_ratio)
figs_1_2_3 <- subplot(fig1, fig2, fig3, nrows = 3, shareX = TRUE) %>% layout(hovermode = "x unified", hoversubplots = "axis")
figs_1_2_3## Warning: 'layout' objects don't have these attributes: 'hoversubplots'
## Valid attributes include:
## '_deprecated', 'activeshape', 'annotations', 'autosize', 'autotypenumbers', 'calendar', 'clickmode', 'coloraxis', 'colorscale', 'colorway', 'computed', 'datarevision', 'dragmode', 'editrevision', 'editType', 'font', 'geo', 'grid', 'height', 'hidesources', 'hoverdistance', 'hoverlabel', 'hovermode', 'images', 'legend', 'mapbox', 'margin', 'meta', 'metasrc', 'modebar', 'newshape', 'paper_bgcolor', 'plot_bgcolor', 'polar', 'scene', 'selectdirection', 'selectionrevision', 'separators', 'shapes', 'showlegend', 'sliders', 'smith', 'spikedistance', 'template', 'ternary', 'title', 'transition', 'uirevision', 'uniformtext', 'updatemenus', 'width', 'xaxis', 'yaxis', 'barmode', 'bargap', 'mapType'To Do: add subplot titles
Now, let’s show price of palladium ETF: PALL
library(tidyquant) # for downloading stock prices## Registered S3 method overwritten by 'quantmod':
## method from
## as.zoo.data.frame zoo## ── Attaching core tidyquant packages ──────────────────────── tidyquant 1.0.9 ──
## ✔ PerformanceAnalytics 2.0.4 ✔ TTR 0.24.4
## ✔ quantmod 0.4.26 ✔ xts 0.14.0
## ── Conflicts ────────────────────────────────────────── tidyquant_conflicts() ──
## ✖ zoo::as.Date() masks base::as.Date()
## ✖ zoo::as.Date.numeric() masks base::as.Date.numeric()
## ✖ plotly::filter() masks dplyr::filter(), stats::filter()
## ✖ xts::first() masks dplyr::first()
## ✖ dplyr::lag() masks stats::lag()
## ✖ xts::last() masks dplyr::last()
## ✖ PerformanceAnalytics::legend() masks graphics::legend()
## ✖ quantmod::summary() masks base::summary()
## ℹ Use the conflicted package (<http://conflicted.r-lib.org/>) to force all conflicts to become errors# or maybe need library(quantmod) ???
# tidyquant includes quantmod
# Add a price of Palladium
# get Pall price
options("getSymbols.warning4.0"=FALSE) # not sure if this is necessary
options("getSymbols.yahoo.warning"=FALSE) # not sure if this is necessary
# Downloading Apple price using quantmod
getSymbols("PALL", from = '2015-01-01',
to = "2017-01-01",warnings = FALSE,
auto.assign = TRUE)## [1] "PALL"head(PALL)## PALL.Open PALL.High PALL.Low PALL.Close PALL.Volume PALL.Adjusted
## 2015-01-02 77.00 77.59 77.00 77.11 10300 77.11
## 2015-01-05 77.20 77.74 76.90 77.09 21900 77.09
## 2015-01-06 77.41 78.54 77.30 77.82 32600 77.82
## 2015-01-07 77.23 77.23 76.25 76.80 49700 76.80
## 2015-01-08 76.91 77.36 76.60 76.76 26400 76.76
## 2015-01-09 77.10 77.91 77.10 77.83 14200 77.83This gives an “xts” object which can be plotted with chart_Series.
chart_Series(PALL)
We can also use this and just one column of data for a simpler plot.
pall_tq <- tq_get('PALL', from = '2015-01-01', to = "2017-01-01", get = "stock.prices" )
head(pall_tq)## # A tibble: 6 × 8
## symbol date open high low close volume adjusted
## <chr> <date> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
## 1 PALL 2015-01-02 77 77.6 77 77.1 10300 77.1
## 2 PALL 2015-01-05 77.2 77.7 76.9 77.1 21900 77.1
## 3 PALL 2015-01-06 77.4 78.5 77.3 77.8 32600 77.8
## 4 PALL 2015-01-07 77.2 77.2 76.2 76.8 49700 76.8
## 5 PALL 2015-01-08 76.9 77.4 76.6 76.8 26400 76.8
## 6 PALL 2015-01-09 77.1 77.9 77.1 77.8 14200 77.8pall_plot <- ggplot(pall_tq, aes(x=date, y = adjusted)) +
geom_line()
pall_plot
This gives us a tibble dataframe.
Let’s try to add the PALL price to the other charts:
fig1 <- ggplotly(comm_long_short)
fig2 <- ggplotly(comm_long_short_ratio)
fig3 <- ggplotly(comm_short_long_ratio)
fig4 <- ggplotly(pall_plot)
figs_1_2_3_4 <- subplot(fig1, fig2, fig3, fig4, nrows = 4, shareX = TRUE) %>% layout(hovermode = "x unified", hoversubplots = "axis")
figs_1_2_3_4## Warning: 'layout' objects don't have these attributes: 'hoversubplots'
## Valid attributes include:
## '_deprecated', 'activeshape', 'annotations', 'autosize', 'autotypenumbers', 'calendar', 'clickmode', 'coloraxis', 'colorscale', 'colorway', 'computed', 'datarevision', 'dragmode', 'editrevision', 'editType', 'font', 'geo', 'grid', 'height', 'hidesources', 'hoverdistance', 'hoverlabel', 'hovermode', 'images', 'legend', 'mapbox', 'margin', 'meta', 'metasrc', 'modebar', 'newshape', 'paper_bgcolor', 'plot_bgcolor', 'polar', 'scene', 'selectdirection', 'selectionrevision', 'separators', 'shapes', 'showlegend', 'sliders', 'smith', 'spikedistance', 'template', 'ternary', 'title', 'transition', 'uirevision', 'uniformtext', 'updatemenus', 'width', 'xaxis', 'yaxis', 'barmode', 'bargap', 'mapType'Clearly, that did not work. Maybe need to add palladium data to the
other tibble dataframe.
Might be as issue with format. One has format “Date” and the other has
format “POSIXct”.
That was the issue. Corrected below. Used:
https://www.statology.org/lubridate-convert-character-to-date/
https://lubridate.tidyverse.org/reference/as_date.html
We then used anti_join function to find rows of palladium_FUT15_16 that
DO NOT have a match in pall_tq.
There were zero. That is good.
We then used semi_join function to find rows of palladium_FUT15_16 that
DO have a match in pall_tq.
There were 104, which is every row in that tibble dataframe. That is
good.
Used:
https://www.listendata.com/2016/08/dplyr-tutorial.html#join_function
# make a new tibble to use for now
palladium_FUT15_16_temp <- palladium_FUT15_16
library(lubridate) # to convert all to same date format
# rename column in palladium_FUT15_15 to match pall_tq
palladium_FUT15_16_temp <- rename(palladium_FUT15_16_temp, date=Report_Date_as_MM_DD_YYYY)
palladium_FUT15_16_temp$date <- ymd(palladium_FUT15_16_temp$date)
#palladium_FUT15_16 <- rename(palladium_FUT15_16, date=Date)
# find where join might work
#palladium_FUT15_16_with_price <-
anti_join(palladium_FUT15_16_temp, pall_tq, by = "date")## # A tibble: 0 × 128
## # ℹ 128 variables: Market_and_Exchange_Names <chr>,
## # As_of_Date_In_Form_YYMMDD <chr>, date <date>,
## # CFTC_Contract_Market_Code <chr>, CFTC_Market_Code <chr>,
## # CFTC_Region_Code <chr>, CFTC_Commodity_Code <chr>, Open_Interest_All <dbl>,
## # NonComm_Positions_Long_All <dbl>, NonComm_Positions_Short_All <dbl>,
## # NonComm_Postions_Spread_All <dbl>, Comm_Positions_Long_All <dbl>,
## # Comm_Positions_Short_All <dbl>, Tot_Rept_Positions_Long_All <dbl>, …semi_join(palladium_FUT15_16_temp, pall_tq, by = "date")## # A tibble: 104 × 128
## Market_and_Exchange_Names As_of_Date_In_Form_YYMMDD date
## <chr> <chr> <date>
## 1 PALLADIUM - NEW YORK MERCANTILE EXCHANGE 161227 2016-12-27
## 2 PALLADIUM - NEW YORK MERCANTILE EXCHANGE 161220 2016-12-20
## 3 PALLADIUM - NEW YORK MERCANTILE EXCHANGE 161213 2016-12-13
## 4 PALLADIUM - NEW YORK MERCANTILE EXCHANGE 161206 2016-12-06
## 5 PALLADIUM - NEW YORK MERCANTILE EXCHANGE 161129 2016-11-29
## 6 PALLADIUM - NEW YORK MERCANTILE EXCHANGE 161122 2016-11-22
## 7 PALLADIUM - NEW YORK MERCANTILE EXCHANGE 161115 2016-11-15
## 8 PALLADIUM - NEW YORK MERCANTILE EXCHANGE 161108 2016-11-08
## 9 PALLADIUM - NEW YORK MERCANTILE EXCHANGE 161101 2016-11-01
## 10 PALLADIUM - NEW YORK MERCANTILE EXCHANGE 161025 2016-10-25
## # ℹ 94 more rows
## # ℹ 125 more variables: CFTC_Contract_Market_Code <chr>,
## # CFTC_Market_Code <chr>, CFTC_Region_Code <chr>, CFTC_Commodity_Code <chr>,
## # Open_Interest_All <dbl>, NonComm_Positions_Long_All <dbl>,
## # NonComm_Positions_Short_All <dbl>, NonComm_Postions_Spread_All <dbl>,
## # Comm_Positions_Long_All <dbl>, Comm_Positions_Short_All <dbl>,
## # Tot_Rept_Positions_Long_All <dbl>, Tot_Rept_Positions_Short_All <dbl>, …Let’s add pall_tq to palladium_FUT15_16 to see if we can plot better.
palladium_FUT15_16_temp_2 <- left_join(palladium_FUT15_16_temp, pall_tq, by = "date")
palladium_FUT15_16_temp_2## # A tibble: 104 × 135
## Market_and_Exchange_Names As_of_Date_In_Form_YYMMDD date
## <chr> <chr> <date>
## 1 PALLADIUM - NEW YORK MERCANTILE EXCHANGE 161227 2016-12-27
## 2 PALLADIUM - NEW YORK MERCANTILE EXCHANGE 161220 2016-12-20
## 3 PALLADIUM - NEW YORK MERCANTILE EXCHANGE 161213 2016-12-13
## 4 PALLADIUM - NEW YORK MERCANTILE EXCHANGE 161206 2016-12-06
## 5 PALLADIUM - NEW YORK MERCANTILE EXCHANGE 161129 2016-11-29
## 6 PALLADIUM - NEW YORK MERCANTILE EXCHANGE 161122 2016-11-22
## 7 PALLADIUM - NEW YORK MERCANTILE EXCHANGE 161115 2016-11-15
## 8 PALLADIUM - NEW YORK MERCANTILE EXCHANGE 161108 2016-11-08
## 9 PALLADIUM - NEW YORK MERCANTILE EXCHANGE 161101 2016-11-01
## 10 PALLADIUM - NEW YORK MERCANTILE EXCHANGE 161025 2016-10-25
## # ℹ 94 more rows
## # ℹ 132 more variables: CFTC_Contract_Market_Code <chr>,
## # CFTC_Market_Code <chr>, CFTC_Region_Code <chr>, CFTC_Commodity_Code <chr>,
## # Open_Interest_All <dbl>, NonComm_Positions_Long_All <dbl>,
## # NonComm_Positions_Short_All <dbl>, NonComm_Postions_Spread_All <dbl>,
## # Comm_Positions_Long_All <dbl>, Comm_Positions_Short_All <dbl>,
## # Tot_Rept_Positions_Long_All <dbl>, Tot_Rept_Positions_Short_All <dbl>, …Seems ok.
Let’s plot “adjusted” price. And with a ratio.
palladium_price <- ggplot(palladium_FUT15_16_temp_2) +
geom_line(aes(x=date,y=adjusted)) +
labs(
y="Ratio",
x="Date",
title="Price of PALL ETF"
)
plot(palladium_price)
comm_short_long_ratio <- ggplot(palladium_FUT15_16) +
geom_line(aes(x=Report_Date_as_MM_DD_YYYY,y=short_long_ratio)) +
labs(
y="Ratio",
x="Date",
title="Ratio of Commercial Short Interest to Commercial Long Interest - Palladium Futures"
)
plot(comm_short_long_ratio)
That works.
Now, back to putting them all together.
# Redo
# Make a plot with palladium data
comm_long_short <- ggplot(palladium_FUT15_16_temp_2) +
# line for Commercial Long
geom_line(aes(x=date,y=Comm_Positions_Long_All, colour = "Long"), size = 1) +
# line for Commercial Short
geom_line(aes(x=date,y=Comm_Positions_Short_All, colour = "Short"), size = 1) +
# better colors and legend
scale_color_manual(name = "Commercial Interest", values = c("Short" = "red", "Long" = "darkgreen")) +
# add label
labs(
y="Commercial Interest",
x="Date",
title = "Commercial Futures Interest for Palladium As Reported by cftc.gov"
)
# Redo long short ratio
comm_long_short_ratio <- ggplot(palladium_FUT15_16_temp_2) +
geom_line(aes(x=date,y=long_short_ratio)) +
labs(
y="Ratio",
x="Date",
title="Ratio of Commercial Long Interest to Commercial Short Interest - Palladium Futures"
)
# Redo short long ratio
comm_short_long_ratio <- ggplot(palladium_FUT15_16_temp_2) +
geom_line(aes(x=date,y=short_long_ratio)) +
labs(
y="Ratio",
x="Date",
title="Ratio of Commercial Short Interest to Commercial Long Interest - Palladium Futures"
)
fig1 <- ggplotly(comm_long_short)
fig2 <- ggplotly(comm_long_short_ratio)
fig3 <- ggplotly(comm_short_long_ratio)
fig4 <- ggplotly(palladium_price)
figs_1_2_3_4 <- subplot(fig1, fig2, fig3, fig4, nrows = 4, shareX = TRUE) %>% layout(hovermode = "x unified", hoversubplots = "axis", title = "Palladium")
figs_1_2_3_4## Warning: 'layout' objects don't have these attributes: 'hoversubplots'
## Valid attributes include:
## '_deprecated', 'activeshape', 'annotations', 'autosize', 'autotypenumbers', 'calendar', 'clickmode', 'coloraxis', 'colorscale', 'colorway', 'computed', 'datarevision', 'dragmode', 'editrevision', 'editType', 'font', 'geo', 'grid', 'height', 'hidesources', 'hoverdistance', 'hoverlabel', 'hovermode', 'images', 'legend', 'mapbox', 'margin', 'meta', 'metasrc', 'modebar', 'newshape', 'paper_bgcolor', 'plot_bgcolor', 'polar', 'scene', 'selectdirection', 'selectionrevision', 'separators', 'shapes', 'showlegend', 'sliders', 'smith', 'spikedistance', 'template', 'ternary', 'title', 'transition', 'uirevision', 'uniformtext', 'updatemenus', 'width', 'xaxis', 'yaxis', 'barmode', 'bargap', 'mapType'That works better. Need to clean up titles and such.
https://plotly.com/r/subplots/.
annotations = list(
list(
x = 0.2,
y = 1.0,
text = "Plot 1",
xref = "paper",
yref = "paper",
xanchor = "center",
yanchor = "bottom",
showarrow = FALSE
),
list(
x = 0.8,
y = 1,
text = "Plot 2",
xref = "paper",
yref = "paper",
xanchor = "center",
yanchor = "bottom",
showarrow = FALSE
),
list(
x = 0.2,
y = 0.4,
text = "Plot 3",
xref = "paper",
yref = "paper",
xanchor = "center",
yanchor = "bottom",
showarrow = FALSE
),
list(
x = 0.8,
y = 0.4,
text = "Plot 4",
xref = "paper",
yref = "paper",
xanchor = "center",
yanchor = "bottom",
showarrow = FALSE
))
figs_1_2_3_4 <- figs_1_2_3_4 %>%layout(annotations = annotations)
figs_1_2_3_4## Warning: 'layout' objects don't have these attributes: 'hoversubplots'
## Valid attributes include:
## '_deprecated', 'activeshape', 'annotations', 'autosize', 'autotypenumbers', 'calendar', 'clickmode', 'coloraxis', 'colorscale', 'colorway', 'computed', 'datarevision', 'dragmode', 'editrevision', 'editType', 'font', 'geo', 'grid', 'height', 'hidesources', 'hoverdistance', 'hoverlabel', 'hovermode', 'images', 'legend', 'mapbox', 'margin', 'meta', 'metasrc', 'modebar', 'newshape', 'paper_bgcolor', 'plot_bgcolor', 'polar', 'scene', 'selectdirection', 'selectionrevision', 'separators', 'shapes', 'showlegend', 'sliders', 'smith', 'spikedistance', 'template', 'ternary', 'title', 'transition', 'uirevision', 'uniformtext', 'updatemenus', 'width', 'xaxis', 'yaxis', 'barmode', 'bargap', 'mapType'Got some stuff working. Now try to reproduce the original plot.
# Make a plot with palladium data
comm_long_short_bar <- ggplot(palladium_FUT15_16_temp_2, aes(x = date)) +
# line for Commercial Long
geom_bar(aes(y=Comm_Positions_Long_All, colour = "Long"), stat = "identity") +
# line for Commercial Short
geom_bar(aes(y=Comm_Positions_Short_All, colour = "Short"), stat = "identity") +
# better colors and legend
scale_color_manual(name = "Commercial Interest", values = c("Short" = "red", "Long" = "darkgreen")) +
# add label
labs(
y="Commercial Interest",
x="Date",
title = "Commercial Futures Interest for Palladium As Reported by cftc.gov"
)
# show plot.
plot(comm_long_short_bar)
Both on same plot overlayed or something. Need side by side or above
and below axis.
For ggplot2 and bar charts, it’s a bit more difficult. Need to rearrange
data.
https://dk81.github.io/dkmathstats_site/rvisual-sidebyside-bar.html
http://www.sthda.com/english/wiki/ggplot2-barplots-quick-start-guide-r-software-and-data-visualization#google_vignette
https://r-graph-gallery.com/48-grouped-barplot-with-ggplot2
https://www.rpubs.com/dvdunne/ggplot_two_bars
https://stackoverflow.com/questions/42820677/ggplot-bar-plot-side-by-side-using-two-variables
# Need to rearrange data for bar plot for geom_bar
palladium_FUT15_16_temp_2_with_pivot <- palladium_FUT15_16_temp_2 %>% pivot_longer(cols = c('Comm_Positions_Long_All','Comm_Positions_Short_All'), names_to = "variable", values_to = "value")
# Make a plot with palladium data
comm_long_short_bar <- ggplot(palladium_FUT15_16_temp_2_with_pivot, aes(x = date, y = value, fill=variable)) +
# line for Commercial Long
geom_bar(stat = "identity", position = "dodge")
# line for Commercial Short
# geom_bar(aes(y=Comm_Positions_Short_All, colour = "Short"), stat = "identity") +
# # better colors and legend
# scale_color_manual(name = "Commercial Interest", values = c("Short" = "red", "Long" = "darkgreen")) +
# # add label
# labs(
# y="Commercial Interest",
# x="Date",
# title = "Commercial Futures Interest for Palladium As Reported by cftc.gov"
#)
# show plot.
plot(comm_long_short_bar)
Hmmm… this is something.
Make short interest negative:
need help here
https://forum.posit.co/t/mutate-with-case-when-and-replace/98315
https://dplyr.tidyverse.org/reference/case_when.html
https://www.tutorialspoint.com/how-to-display-negative-labels-below-bars-in-barplot-using-ggplot2-in-r
https://r-graph-gallery.com/341-stacked-barplot-with-negative-values.html
# Need to rearrange data for bar plot for geom_bar
palladium_FUT15_16_temp_2_with_pivot <- palladium_FUT15_16_temp_2_with_pivot %>%
mutate(value = ifelse(variable =="Comm_Positions_Short_All", value * -1, value))
# Make a plot with palladium data
comm_long_short_bar <- ggplot(palladium_FUT15_16_temp_2_with_pivot, aes(x = date, y = value, fill=variable)) +
# line for Commercial Long
geom_bar(stat = "identity", position = "dodge")
# line for Commercial Short
# geom_bar(aes(y=Comm_Positions_Short_All, colour = "Short"), stat = "identity") +
# # better colors and legend
# scale_color_manual(name = "Commercial Interest", values = c("Short" = "red", "Long" = "darkgreen")) +
# # add label
# labs(
# y="Commercial Interest",
# x="Date",
# title = "Commercial Futures Interest for Palladium As Reported by cftc.gov"
#)
# show plot.
plot(comm_long_short_bar)
Now with plotly:
Multiple chart types like example:
https://plotly.com/r/graphing-multiple-chart-types/
Time series and financial data:
https://plotly.com/r/time-series/#hover-templates-with-mixtures-of-period-data
ggplotly(comm_long_short_bar)That works.
Clean up colors and axis. Add trace for both.
fig1 <- ggplotly(comm_long_short_bar)
fig2 <- ggplotly(comm_long_short_ratio)
fig3 <- ggplotly(comm_short_long_ratio)
fig4 <- ggplotly(palladium_price)
figs_1_2_3_4 <- subplot(fig1, fig2, fig3, fig4, nrows = 4, shareX = TRUE) %>% layout(hovermode = "x unified", hoversubplots = "axis", title = "Palladium")
figs_1_2_3_4## Warning: 'layout' objects don't have these attributes: 'hoversubplots'
## Valid attributes include:
## '_deprecated', 'activeshape', 'annotations', 'autosize', 'autotypenumbers', 'calendar', 'clickmode', 'coloraxis', 'colorscale', 'colorway', 'computed', 'datarevision', 'dragmode', 'editrevision', 'editType', 'font', 'geo', 'grid', 'height', 'hidesources', 'hoverdistance', 'hoverlabel', 'hovermode', 'images', 'legend', 'mapbox', 'margin', 'meta', 'metasrc', 'modebar', 'newshape', 'paper_bgcolor', 'plot_bgcolor', 'polar', 'scene', 'selectdirection', 'selectionrevision', 'separators', 'shapes', 'showlegend', 'sliders', 'smith', 'spikedistance', 'template', 'ternary', 'title', 'transition', 'uirevision', 'uniformtext', 'updatemenus', 'width', 'xaxis', 'yaxis', 'barmode', 'bargap', 'mapType'Try to max and min of next x rows for use for max gain and max loss
in next xxx months.
Make new columns via mutate and take a look.
https://stackoverflow.com/questions/66049527/r-display-first-10-rows-of-data-of-certain-columns
slider library for getting max or min value in last xxx rows
https://cran.r-project.org/web/packages/slider/vignettes/slider.html
https://www.tidyverse.org/blog/2020/02/slider-0-1-0/
https://stackoverflow.com/questions/38405025/find-running-minimum-and-max-in-r
https://stackoverflow.com/questions/63632648/how-to-find-the-maximum-value-in-the-20-previous-rows-of-a-column-in-r
library(slider)
# for now using without pivot
palladium_FUT15_16_temp_2_max_and_min <- palladium_FUT15_16_temp_2 %>%
mutate(six_month_max = slide_dbl(adjusted, max, .before = 26, .complete = TRUE)) %>%
mutate(six_month_min = slide_dbl(adjusted, min, .before = 26, .complete = TRUE)) %>%
mutate(max_gain = 100*(six_month_max-adjusted)/adjusted) %>%
mutate(max_loss = 100*(six_month_min-adjusted)/adjusted)
head(palladium_FUT15_16_temp_2_max_and_min,20)[c("adjusted","six_month_max","six_month_min","max_gain","max_loss")]## # A tibble: 20 × 5
## adjusted six_month_max six_month_min max_gain max_loss
## <dbl> <dbl> <dbl> <dbl> <dbl>
## 1 64.6 NA NA NA NA
## 2 64.0 NA NA NA NA
## 3 70.0 NA NA NA NA
## 4 70.5 NA NA NA NA
## 5 73.2 NA NA NA NA
## 6 71.3 NA NA NA NA
## 7 67.8 NA NA NA NA
## 8 63.6 NA NA NA NA
## 9 60.8 NA NA NA NA
## 10 60.8 NA NA NA NA
## 11 61.4 NA NA NA NA
## 12 62.2 NA NA NA NA
## 13 67.0 NA NA NA NA
## 14 67.2 NA NA NA NA
## 15 65.6 NA NA NA NA
## 16 63.0 NA NA NA NA
## 17 67.1 NA NA NA NA
## 18 65.3 NA NA NA NA
## 19 67.2 NA NA NA NA
## 20 67.5 NA NA NA NALet’s try to look at plots of max gain and loss against price and long/short ratios.
comm_long_short_maxmin <- ggplot(palladium_FUT15_16_temp_2_max_and_min) +
# line for Commercial Long
geom_line(aes(x=date,y=Comm_Positions_Long_All, colour = "Long"), size = 1) +
# line for Commercial Short
geom_line(aes(x=date,y=Comm_Positions_Short_All, colour = "Short"), size = 1) +
# better colors and legend
scale_color_manual(name = "Commercial Interest", values = c("Short" = "red", "Long" = "darkgreen")) +
# add label
labs(
y="Commercial Interest",
x="Date",
title = "Commercial Futures Interest for Palladium As Reported by cftc.gov"
)
# Redo long short ratio
comm_long_short_ratio_maxmin <- ggplot(palladium_FUT15_16_temp_2_max_and_min) +
geom_line(aes(x=date,y=long_short_ratio)) +
labs(
y="Ratio",
x="Date",
title="Ratio of Commercial Long Interest to Commercial Short Interest - Palladium Futures"
)
# Redo short long ratio
comm_short_long_ratio_maxmin <- ggplot(palladium_FUT15_16_temp_2_max_and_min) +
geom_line(aes(x=date,y=short_long_ratio)) +
labs(
y="Ratio",
x="Date",
title="Ratio of Commercial Short Interest to Commercial Long Interest - Palladium Futures"
)
palladium_price_maxmin <- ggplot(palladium_FUT15_16_temp_2_max_and_min) +
geom_line(aes(x=date,y=adjusted)) +
labs(
y="Ratio",
x="Date",
title="Price of PALL ETF"
)
#plot(palladium_price)
max_gain_maxmin <- ggplot(palladium_FUT15_16_temp_2_max_and_min) +
geom_line(aes(x=date,y=max_gain))
max_loss_maxmin <- ggplot(palladium_FUT15_16_temp_2_max_and_min) +
geom_line(aes(x=date,y=max_loss))
fig1 <- ggplotly(comm_long_short_maxmin)
fig2 <- ggplotly(comm_long_short_ratio_maxmin)
fig3 <- ggplotly(comm_short_long_ratio_maxmin)
fig4 <- ggplotly(palladium_price_maxmin)
fig5 <- ggplotly(max_gain_maxmin)
fig6 <- ggplotly(max_loss_maxmin)
figs_1_2_3_4_5_5_maxmin <- subplot(fig1, fig2, fig3, fig4, fig5, fig6, nrows = 6, shareX = TRUE) %>% layout(hovermode = "x unified", hoversubplots = "axis", title = "Palladium")
figs_1_2_3_4_5_5_maxmin## Warning: 'layout' objects don't have these attributes: 'hoversubplots'
## Valid attributes include:
## '_deprecated', 'activeshape', 'annotations', 'autosize', 'autotypenumbers', 'calendar', 'clickmode', 'coloraxis', 'colorscale', 'colorway', 'computed', 'datarevision', 'dragmode', 'editrevision', 'editType', 'font', 'geo', 'grid', 'height', 'hidesources', 'hoverdistance', 'hoverlabel', 'hovermode', 'images', 'legend', 'mapbox', 'margin', 'meta', 'metasrc', 'modebar', 'newshape', 'paper_bgcolor', 'plot_bgcolor', 'polar', 'scene', 'selectdirection', 'selectionrevision', 'separators', 'shapes', 'showlegend', 'sliders', 'smith', 'spikedistance', 'template', 'ternary', 'title', 'transition', 'uirevision', 'uniformtext', 'updatemenus', 'width', 'xaxis', 'yaxis', 'barmode', 'bargap', 'mapType'You can always zoom in plotly by selecting a range right on the plot.
add a rangeslider for date
https://plotly-r.com/animating-views
https://plotly.com/r/range-slider/
https://plotly.com/r/reference/layout/sliders/#layout-sliders
https://stackoverflow.com/questions/70943105/how-to-dynamically-change-plotly-axis-based-on-crosstalk-conditions/
https://plot.ly/r/sliders/
comm_long_short_maxmin <- ggplot(palladium_FUT15_16_temp_2_max_and_min) +
# line for Commercial Long
geom_line(aes(x=date,y=Comm_Positions_Long_All, colour = "Long"), size = 1) +
# line for Commercial Short
geom_line(aes(x=date,y=Comm_Positions_Short_All, colour = "Short"), size = 1) +
# better colors and legend
scale_color_manual(name = "Commercial Interest", values = c("Short" = "red", "Long" = "darkgreen")) +
# add label
labs(
y="Commercial Interest",
x="Date",
title = "Commercial Futures Interest for Palladium As Reported by cftc.gov"
)
# Redo long short ratio
comm_long_short_ratio_maxmin <- ggplot(palladium_FUT15_16_temp_2_max_and_min) +
geom_line(aes(x=date,y=long_short_ratio)) +
labs(
y="Ratio",
x="Date",
title="Ratio of Commercial Long Interest to Commercial Short Interest - Palladium Futures"
)
# Redo short long ratio
comm_short_long_ratio_maxmin <- ggplot(palladium_FUT15_16_temp_2_max_and_min) +
geom_line(aes(x=date,y=short_long_ratio)) +
labs(
y="Ratio",
x="Date",
title="Ratio of Commercial Short Interest to Commercial Long Interest - Palladium Futures"
)
palladium_price_maxmin <- ggplot(palladium_FUT15_16_temp_2_max_and_min) +
geom_line(aes(x=date,y=adjusted)) +
labs(
y="Ratio",
x="Date",
title="Price of PALL ETF"
)
#plot(palladium_price)
max_gain_maxmin <- ggplot(palladium_FUT15_16_temp_2_max_and_min) +
geom_line(aes(x=date,y=max_gain))
max_loss_maxmin <- ggplot(palladium_FUT15_16_temp_2_max_and_min) +
geom_line(aes(x=date,y=max_loss))
fig1 <- ggplotly(comm_long_short_maxmin) %>% layout(yaxis = list(title = "yaxis test tile"))
fig2 <- ggplotly(comm_long_short_ratio_maxmin)
fig3 <- ggplotly(comm_short_long_ratio_maxmin)
fig4 <- ggplotly(palladium_price_maxmin)
fig5 <- ggplotly(max_gain_maxmin)
fig6 <- ggplotly(max_loss_maxmin)
figs_1_2_3_4_5_5_maxmin <- subplot(fig1, fig2, fig3, fig4, fig5, fig6, nrows = 6, shareX = TRUE) %>% layout(hovermode = "x unified", hoversubplots = "axis", title = "Palladium", xaxis = list(
rangeslider = list(type = "date")))
figs_1_2_3_4_5_5_maxmin## Warning: 'layout' objects don't have these attributes: 'hoversubplots'
## Valid attributes include:
## '_deprecated', 'activeshape', 'annotations', 'autosize', 'autotypenumbers', 'calendar', 'clickmode', 'coloraxis', 'colorscale', 'colorway', 'computed', 'datarevision', 'dragmode', 'editrevision', 'editType', 'font', 'geo', 'grid', 'height', 'hidesources', 'hoverdistance', 'hoverlabel', 'hovermode', 'images', 'legend', 'mapbox', 'margin', 'meta', 'metasrc', 'modebar', 'newshape', 'paper_bgcolor', 'plot_bgcolor', 'polar', 'scene', 'selectdirection', 'selectionrevision', 'separators', 'shapes', 'showlegend', 'sliders', 'smith', 'spikedistance', 'template', 'ternary', 'title', 'transition', 'uirevision', 'uniformtext', 'updatemenus', 'width', 'xaxis', 'yaxis', 'barmode', 'bargap', 'mapType'# layout(
# xaxis = list(
# rangeslider = list(type = "Report_Date_as_MM_DD_YYY")))This works.
#### The plots at the range slider are all supposed to be there.
Now update sizes/heights and add y axis titles.
https://plotly-r.com/arranging-views
https://plotly.com/r/subplots/
Having some issues with getting plot to look how it should
Trying some dummy data first
Make three tibbles and combine like you need to do for the data above.
# Make three dummy tibbles
df1 <- data.frame(date_temp = c(3, 4, 5), comm_long_temp = c(1000, 1100, 1200))
df1 <- as_tibble(df1)
df2 <- data.frame(date_temp = c(2, 3, 4), ratio1_temp = c(100, 110, 120))
df2 <- as_tibble(df2)
df3 <- data.frame(date_temp = c(0, 1, 2), ratio2_temp = c(10, 11, 12))
df3 <- as_tibble(df3)
# join into one tibble
df_max <- full_join(df1, df2, by = "date_temp")
df_max## # A tibble: 4 × 3
## date_temp comm_long_temp ratio1_temp
## <dbl> <dbl> <dbl>
## 1 3 1000 110
## 2 4 1100 120
## 3 5 1200 NA
## 4 2 NA 100df_max <- full_join(df_max, df3, by = "date_temp")
df_max## # A tibble: 6 × 4
## date_temp comm_long_temp ratio1_temp ratio2_temp
## <dbl> <dbl> <dbl> <dbl>
## 1 3 1000 110 NA
## 2 4 1100 120 NA
## 3 5 1200 NA NA
## 4 2 NA 100 12
## 5 0 NA NA 10
## 6 1 NA NA 11We now have a tibble which is all “linked” by date_temp.
Make plots.
# Make three dummy tibbles
df1 <- data.frame(date_temp = c(3, 4, 5), comm_long_temp = c(1000, 1100, 1200))
df1 <- as_tibble(df1)
df2 <- data.frame(date_temp = c(2, 3, 4), ratio1_temp = c(100, 110, 120))
df2 <- as_tibble(df2)
df3 <- data.frame(date_temp = c(0, 1, 2), ratio2_temp = c(10, 11, 12))
df3 <- as_tibble(df3)
# join into one tibble
df_max <- full_join(df1, df2, by = "date_temp")
df_max <- full_join(df_max, df3, by = "date_temp")
# make 3 plots
p1 <- ggplot(df_max) +
geom_line(aes(x = date_temp, y = comm_long_temp))
plot(p1)## Warning: Removed 3 rows containing missing values or values outside the scale range
## (`geom_line()`).
p2 <- ggplot(df_max) +
geom_line(aes(x = date_temp, y = ratio1_temp))
plot(p2)## Warning: Removed 3 rows containing missing values or values outside the scale range
## (`geom_line()`).
p3 <- ggplot(df_max) +
geom_line(aes(x = date_temp, y = ratio2_temp))
plot(p3)## Warning: Removed 3 rows containing missing values or values outside the scale range
## (`geom_line()`).
Now make three plotly plots.
# Make three dummy tibbles
df1 <- data.frame(date_temp = c(3, 4, 5), comm_long_temp = c(1000, 1100, 1200))
df1 <- as_tibble(df1)
df2 <- data.frame(date_temp = c(2, 3, 4), ratio1_temp = c(100, 110, 120))
df2 <- as_tibble(df2)
df3 <- data.frame(date_temp = c(0, 1, 2), ratio2_temp = c(10, 11, 12))
df3 <- as_tibble(df3)
# join into one tibble
df_max <- full_join(df1, df2, by = "date_temp")
df_max <- full_join(df_max, df3, by = "date_temp")
# make 3 plotly plots
p1 <- ggplot(df_max) +
geom_line(aes(x = date_temp, y = comm_long_temp))
#plot(p1)
p1 <- ggplotly(p1)
p1p2 <- ggplot(df_max) +
geom_line(aes(x = date_temp, y = ratio1_temp))
#plot(p2)
p2 <- ggplotly(p2)
p2p3 <- ggplot(df_max) +
geom_line(aes(x = date_temp, y = ratio2_temp))
#plot(p3)
p3 <- ggplotly(p3)
p3Now make subplots on one x axis with same hover line across all plots.
# Make three dummy tibbles
df1 <- data.frame(date_temp = c(3, 4, 5), comm_long_temp = c(1000, 1100, 1200))
df1 <- as_tibble(df1)
df2 <- data.frame(date_temp = c(2, 3, 4), ratio1_temp = c(100, 110, 120))
df2 <- as_tibble(df2)
df3 <- data.frame(date_temp = c(0, 1, 2), ratio2_temp = c(10, 11, 12))
df3 <- as_tibble(df3)
# join into one tibble
df_max <- full_join(df1, df2, by = "date_temp")
df_max <- full_join(df_max, df3, by = "date_temp")
# make 3 plotly plots
p1 <- ggplot(df_max) +
geom_line(aes(x = date_temp, y = comm_long_temp))
#plot(p1)
p1 <- ggplotly(p1)
#p1
p2 <- ggplot(df_max) +
geom_line(aes(x = date_temp, y = ratio1_temp))
#plot(p2)
p2 <- ggplotly(p2)
#p2
p3 <- ggplot(df_max) +
geom_line(aes(x = date_temp, y = ratio2_temp))
#plot(p3)
p3 <- ggplotly(p3)
#p3
# combine on one chart with shared X axis
fig_temp <- subplot(p1, p2, p3, nrows = 3, shareX = TRUE) %>% layout(hovermode = "x unified", hoversubplots = "axis", title = "Palladium")
fig_temp## Warning: 'layout' objects don't have these attributes: 'hoversubplots'
## Valid attributes include:
## '_deprecated', 'activeshape', 'annotations', 'autosize', 'autotypenumbers', 'calendar', 'clickmode', 'coloraxis', 'colorscale', 'colorway', 'computed', 'datarevision', 'dragmode', 'editrevision', 'editType', 'font', 'geo', 'grid', 'height', 'hidesources', 'hoverdistance', 'hoverlabel', 'hovermode', 'images', 'legend', 'mapbox', 'margin', 'meta', 'metasrc', 'modebar', 'newshape', 'paper_bgcolor', 'plot_bgcolor', 'polar', 'scene', 'selectdirection', 'selectionrevision', 'separators', 'shapes', 'showlegend', 'sliders', 'smith', 'spikedistance', 'template', 'ternary', 'title', 'transition', 'uirevision', 'uniformtext', 'updatemenus', 'width', 'xaxis', 'yaxis', 'barmode', 'bargap', 'mapType'Figure out a legend.
https://ggplot2.tidyverse.org/reference/guide_legend.html
https://stackoverflow.com/questions/74392543/how-to-add-a-legend-to-ggplot-in-r
https://r-charts.com/ggplot2/legend/
https://www.tidyverse.org/blog/2024/02/ggplot2-3-5-0-legends/
https://r-graph-gallery.com/239-custom-layout-legend-ggplot2.html
http://www.sthda.com/english/wiki/ggplot2-legend-easy-steps-to-change-the-position-and-the-appearance-of-a-graph-legend-in-r-software
# Make three dummy tibbles
df1 <- data.frame(date_temp = c(3, 4, 5), comm_long_temp = c(1000, 1100, 1200))
df1 <- as_tibble(df1)
df2 <- data.frame(date_temp = c(2, 3, 4), ratio1_temp = c(100, 110, 120))
df2 <- as_tibble(df2)
df3 <- data.frame(date_temp = c(0, 1, 2), ratio2_temp = c(10, 11, 12))
df3 <- as_tibble(df3)
# join into one tibble
df_max <- full_join(df1, df2, by = "date_temp")
df_max <- full_join(df_max, df3, by = "date_temp")
# make 3 plotly plots
p1 <- ggplot(df_max) +
geom_line(aes(x = date_temp, y = comm_long_temp, colour = "comm_long_temp"))
#plot(p1)
p1 <- ggplotly(p1)
#p1
p2 <- ggplot(df_max) +
geom_line(aes(x = date_temp, y = ratio1_temp, colour = "ratio1_temp"))
#plot(p2)
p2 <- ggplotly(p2)
#p2
p3 <- ggplot(df_max) +
geom_line(aes(x = date_temp, y = ratio2_temp, colour = "ratio2_temp"))
#plot(p3)
p3 <- ggplotly(p3)
#p3
# combine on one chart with shared X axis
fig_temp <- subplot(p1, p2, p3, nrows = 3, shareX = TRUE) %>% layout(hovermode = "x unified", hoversubplots = "axis", title = "Palladium")
fig_temp## Warning: 'layout' objects don't have these attributes: 'hoversubplots'
## Valid attributes include:
## '_deprecated', 'activeshape', 'annotations', 'autosize', 'autotypenumbers', 'calendar', 'clickmode', 'coloraxis', 'colorscale', 'colorway', 'computed', 'datarevision', 'dragmode', 'editrevision', 'editType', 'font', 'geo', 'grid', 'height', 'hidesources', 'hoverdistance', 'hoverlabel', 'hovermode', 'images', 'legend', 'mapbox', 'margin', 'meta', 'metasrc', 'modebar', 'newshape', 'paper_bgcolor', 'plot_bgcolor', 'polar', 'scene', 'selectdirection', 'selectionrevision', 'separators', 'shapes', 'showlegend', 'sliders', 'smith', 'spikedistance', 'template', 'ternary', 'title', 'transition', 'uirevision', 'uniformtext', 'updatemenus', 'width', 'xaxis', 'yaxis', 'barmode', 'bargap', 'mapType'change name of legend
# Make three dummy tibbles
df1 <- data.frame(date_temp = c(3, 4, 5), comm_long_temp = c(1000, 1100, 1200))
df1 <- as_tibble(df1)
df2 <- data.frame(date_temp = c(2, 3, 4), ratio1_temp = c(100, 110, 120))
df2 <- as_tibble(df2)
df3 <- data.frame(date_temp = c(0, 1, 2), ratio2_temp = c(10, 11, 12))
df3 <- as_tibble(df3)
# join into one tibble
df_max <- full_join(df1, df2, by = "date_temp")
df_max <- full_join(df_max, df3, by = "date_temp")
# make 3 plotly plots
p1 <- ggplot(df_max) +
geom_line(aes(x = date_temp, y = comm_long_temp, colour = "comm_long_temp")) +
scale_color_manual(values = "red") +
labs(colour = "test")
#plot(p1)
p1 <- ggplotly(p1)
#p1
p2 <- ggplot(df_max) +
geom_line(aes(x = date_temp, y = ratio1_temp, colour = "ratio1_temp")) +
scale_color_manual(values = "green") +
labs(colour = "test")
#plot(p2)
p2 <- ggplotly(p2)
#p2
p3 <- ggplot(df_max) +
geom_line(aes(x = date_temp, y = ratio2_temp, colour = "ratio2_temp")) +
scale_color_manual(values = "blue") +
labs(colour = "test")
#plot(p3)
p3 <- ggplotly(p3)
#p3
# combine on one chart with shared X axis
fig_temp <- subplot(p1, p2, p3, nrows = 3, shareX = TRUE) %>% layout(hovermode = "x unified", hoversubplots = "axis", title = "Palladium")
fig_temp## Warning: 'layout' objects don't have these attributes: 'hoversubplots'
## Valid attributes include:
## '_deprecated', 'activeshape', 'annotations', 'autosize', 'autotypenumbers', 'calendar', 'clickmode', 'coloraxis', 'colorscale', 'colorway', 'computed', 'datarevision', 'dragmode', 'editrevision', 'editType', 'font', 'geo', 'grid', 'height', 'hidesources', 'hoverdistance', 'hoverlabel', 'hovermode', 'images', 'legend', 'mapbox', 'margin', 'meta', 'metasrc', 'modebar', 'newshape', 'paper_bgcolor', 'plot_bgcolor', 'polar', 'scene', 'selectdirection', 'selectionrevision', 'separators', 'shapes', 'showlegend', 'sliders', 'smith', 'spikedistance', 'template', 'ternary', 'title', 'transition', 'uirevision', 'uniformtext', 'updatemenus', 'width', 'xaxis', 'yaxis', 'barmode', 'bargap', 'mapType'change name of individual lines
# Make three dummy tibbles
df1 <- data.frame(date_temp = c(3, 4, 5), comm_long_temp = c(1000, 1100, 1200))
df1 <- as_tibble(df1)
df2 <- data.frame(date_temp = c(2, 3, 4), ratio1_temp = c(100, 110, 120))
df2 <- as_tibble(df2)
df3 <- data.frame(date_temp = c(0, 1, 2), ratio2_temp = c(10, 11, 12))
df3 <- as_tibble(df3)
# join into one tibble
df_max <- full_join(df1, df2, by = "date_temp")
df_max <- full_join(df_max, df3, by = "date_temp")
# make 3 plotly plots
p1 <- ggplot(df_max) +
geom_line(aes(x = date_temp, y = comm_long_temp, colour = "Commercial Long COT")) +
scale_color_manual(values = "red") +
labs(colour = "Commercials")
#plot(p1)
p1 <- ggplotly(p1)
#p1
p2 <- ggplot(df_max) +
geom_line(aes(x = date_temp, y = ratio1_temp, colour = "Commercial Ratio Long")) +
scale_color_manual(values = "green") +
labs(colour = "Commercials")
#plot(p2)
p2 <- ggplotly(p2)
#p2
p3 <- ggplot(df_max) +
geom_line(aes(x = date_temp, y = ratio2_temp, colour = "Commercial Ratio Short")) +
scale_color_manual(values = "blue") +
labs(colour = "Commercials")
#plot(p3)
p3 <- ggplotly(p3)
#p3
# combine on one chart with shared X axis
fig_temp <- subplot(p1, p2, p3, nrows = 3, shareX = TRUE) %>% layout(hovermode = "x unified", hoversubplots = "axis", title = "Palladium")
fig_temp## Warning: 'layout' objects don't have these attributes: 'hoversubplots'
## Valid attributes include:
## '_deprecated', 'activeshape', 'annotations', 'autosize', 'autotypenumbers', 'calendar', 'clickmode', 'coloraxis', 'colorscale', 'colorway', 'computed', 'datarevision', 'dragmode', 'editrevision', 'editType', 'font', 'geo', 'grid', 'height', 'hidesources', 'hoverdistance', 'hoverlabel', 'hovermode', 'images', 'legend', 'mapbox', 'margin', 'meta', 'metasrc', 'modebar', 'newshape', 'paper_bgcolor', 'plot_bgcolor', 'polar', 'scene', 'selectdirection', 'selectionrevision', 'separators', 'shapes', 'showlegend', 'sliders', 'smith', 'spikedistance', 'template', 'ternary', 'title', 'transition', 'uirevision', 'uniformtext', 'updatemenus', 'width', 'xaxis', 'yaxis', 'barmode', 'bargap', 'mapType'add points and those colors.
http://www.sthda.com/english/wiki/ggplot2-line-plot-quick-start-guide-r-software-and-data-visualization
https://www.appsilon.com/post/ggplot2-line-charts
https://stackoverflow.com/questions/27350243/ggplot-line-graph-with-different-line-styles-and-markers
https://stackoverflow.com/questions/31069324/adding-points-from-other-dataset-to-ggplot2
# Make three dummy tibbles
df1 <- data.frame(date_temp = c(3, 4, 5), comm_long_temp = c(1000, 1100, 1200))
df1 <- as_tibble(df1)
df2 <- data.frame(date_temp = c(2, 3, 4), ratio1_temp = c(100, 110, 120))
df2 <- as_tibble(df2)
df3 <- data.frame(date_temp = c(0, 1, 2), ratio2_temp = c(10, 11, 12))
df3 <- as_tibble(df3)
# join into one tibble
df_max <- full_join(df1, df2, by = "date_temp")
df_max <- full_join(df_max, df3, by = "date_temp")
# make 3 plotly plots
p1 <- ggplot(df_max) +
geom_line(aes(x = date_temp, y = comm_long_temp, colour = "Commercial Long COT")) +
geom_point(aes(x = date_temp, y = comm_long_temp, colour = "Commercial Long COT")) +
scale_color_manual(values = "red") +
labs(colour = "Commercials")
#plot(p1)
p1 <- ggplotly(p1)
#p1
p2 <- ggplot(df_max) +
geom_line(aes(x = date_temp, y = ratio1_temp, colour = "Commercial Ratio Long")) +
geom_point(aes(x = date_temp, y = ratio1_temp, colour = "Commercial Ratio Long")) +
scale_color_manual(values = "green") +
labs(colour = "Commercials")
#plot(p2)
p2 <- ggplotly(p2)
#p2
p3 <- ggplot(df_max) +
geom_line(aes(x = date_temp, y = ratio2_temp, colour = "Commercial Ratio Short")) +
geom_point(aes(x = date_temp, y = ratio2_temp, colour = "Commercial Ratio Short")) +
scale_color_manual(values = "blue") +
labs(colour = "Commercials")
#plot(p3)
p3 <- ggplotly(p3)
#p3
# combine on one chart with shared X axis
fig_temp <- subplot(p1, p2, p3, nrows = 3, shareX = TRUE) %>% layout(hovermode = "x unified", hoversubplots = "axis", title = "Palladium")
fig_temp## Warning: 'layout' objects don't have these attributes: 'hoversubplots'
## Valid attributes include:
## '_deprecated', 'activeshape', 'annotations', 'autosize', 'autotypenumbers', 'calendar', 'clickmode', 'coloraxis', 'colorscale', 'colorway', 'computed', 'datarevision', 'dragmode', 'editrevision', 'editType', 'font', 'geo', 'grid', 'height', 'hidesources', 'hoverdistance', 'hoverlabel', 'hovermode', 'images', 'legend', 'mapbox', 'margin', 'meta', 'metasrc', 'modebar', 'newshape', 'paper_bgcolor', 'plot_bgcolor', 'polar', 'scene', 'selectdirection', 'selectionrevision', 'separators', 'shapes', 'showlegend', 'sliders', 'smith', 'spikedistance', 'template', 'ternary', 'title', 'transition', 'uirevision', 'uniformtext', 'updatemenus', 'width', 'xaxis', 'yaxis', 'barmode', 'bargap', 'mapType'now add y axis labels.
and change x axis label
library(gridExtra)##
## Attaching package: 'gridExtra'## The following object is masked from 'package:dplyr':
##
## combine# Make three dummy tibbles
df1 <- data.frame(date_temp = c(3, 4, 5), comm_long_temp = c(1000, 1100, 1200))
df1 <- as_tibble(df1)
df2 <- data.frame(date_temp = c(2, 3, 4), ratio1_temp = c(100, 110, 120))
df2 <- as_tibble(df2)
df3 <- data.frame(date_temp = c(0, 1, 2), ratio2_temp = c(10, 11, 12))
df3 <- as_tibble(df3)
# join into one tibble
df_max <- full_join(df1, df2, by = "date_temp")
df_max <- full_join(df_max, df3, by = "date_temp")
# make 3 plotly plots
p1 <- ggplot(df_max) +
geom_line(aes(x = date_temp, y = comm_long_temp, colour = "Commercial Long COT")) +
# geom_point(aes(x = date_temp, y = comm_long_temp, colour = "Commercial Long COT")) +
# scale_color_manual(values = "red") +
# labs(colour = "Commercials", x="Date", y="Ratio")
xlab("p1x") + ylab("p1y")
#plot(p1)
p1 <- ggplotly(p1)
#p1
p2 <- ggplot(df_max) +
geom_line(aes(x = date_temp, y = ratio1_temp, colour = "Commercial Ratio Long")) +
# geom_point(aes(x = date_temp, y = ratio1_temp, colour = "Commercial Ratio Long")) +
# scale_color_manual(values = "green") +
# #labs(colour = "Commercials", x="Date", y="Ratio")
# xlab("sdfs") + ylab("asdfsadf")
xlab("p2x") + ylab("p2y")
#plot(p2)
p2 <- ggplotly(p2)
#p2
p3 <- ggplot(df_max) +
geom_line(aes(x = date_temp, y = ratio2_temp, colour = "Commercial Ratio Short")) +
# geom_point(aes(x = date_temp, y = ratio2_temp, colour = "Commercial Ratio Short")) +
# scale_color_manual(values = "blue") +
# labs(colour = "Commercials", x="Date", y="Ratio")
xlab("p3x") + ylab("p3y")
#plot(p3)
p3 <- ggplotly(p3) #%>%
#layout(xaxis = list(title = "Date"), yaxis = list(title = "Contracts"))
#p3
# combine on one chart with shared X axis
# fig_temp <- subplot(p1, p2, p3, nrows = 3, shareX = TRUE) %>% layout(hovermode = "x unified", hoversubplots = "axis", titleY = TRUE, titleX = TRUE, margin = 0.05)
fig_temp <- subplot(p1, p2, p3, nrows = 3, titleY = TRUE, titleX = TRUE, margin = 0.05)
annotations = list(
list(
x = 0.2,
y = 1.0,
text = "Plot 1",
xref = "paper",
yref = "paper",
xanchor = "center",
yanchor = "bottom",
showarrow = FALSE
),
list(
x = 0.8,
y = 1,
text = "Plot 2",
xref = "paper",
yref = "paper",
xanchor = "center",
yanchor = "bottom",
showarrow = FALSE
),
list(
x = 0.2,
y = 0.4,
text = "Plot 3",
xref = "paper",
yref = "paper",
xanchor = "center",
yanchor = "bottom",
showarrow = FALSE
))
#fig_temp <- fig_temp %>%layout(annotations = annotations)
#fig
fig_tempThe issue is Share X trying to make axis labels show up.
library(ggplot2)
library(gridExtra)
library(plotly)
d <- data.frame(x=1:20,y=1:20, z=20:1)
p1 <- ggplot(data=d) +
geom_point(aes(x=x, y=y)) +
xlab("X") + ylab("Y")
p2 <- ggplot(data=d) +
geom_point(aes(x=x, y=z)) +
xlab("X") + ylab("Z")
ply1 <- ggplotly(p1)
ply2 <- ggplotly(p2)
ply_total <- subplot(ply1, ply2, nrows=1,
titleY = TRUE,
titleX = TRUE,
margin = 0.05)
ply_total <- ply_total %>% layout(autosize = F, width = 500, height = 500)## Warning: Specifying width/height in layout() is now deprecated.
## Please specify in ggplotly() or plot_ly()ply_totalhttps://github.com/plotly/plotly.R/issues/1415
https://stackoverflow.com/questions/75132414/axis-labels-lost-in-plotlysubplot
https://stackoverflow.com/questions/61553282/how-to-correctly-place-axis-labels-in-plotly-r-subplots-with-multiple-axes
https://community.plotly.com/t/subplot-y-axis-names/2683
library(gridExtra)
# Make three dummy tibbles
df1 <- data.frame(date_temp = c(3, 4, 5), comm_long_temp = c(1000, 1100, 1200))
df1 <- as_tibble(df1)
df2 <- data.frame(date_temp = c(2, 3, 4), ratio1_temp = c(100, 110, 120))
df2 <- as_tibble(df2)
df3 <- data.frame(date_temp = c(0, 1, 2), ratio2_temp = c(10, 11, 12))
df3 <- as_tibble(df3)
# join into one tibble
df_max <- full_join(df1, df2, by = "date_temp")
df_max <- full_join(df_max, df3, by = "date_temp")
# make 3 plotly plots
p1 <- ggplot(df_max) +
geom_line(aes(x = date_temp, y = comm_long_temp, colour = "Commercial Long COT")) +
# geom_point(aes(x = date_temp, y = comm_long_temp, colour = "Commercial Long COT")) +
# scale_color_manual(values = "red") +
# labs(colour = "Commercials", x="Date", y="Ratio")
xlab("p1x") + ylab("p1y")
#plot(p1)
p1 <- ggplotly(p1)
#p1
p2 <- ggplot(df_max) +
geom_line(aes(x = date_temp, y = ratio1_temp, colour = "Commercial Ratio Long")) +
# geom_point(aes(x = date_temp, y = ratio1_temp, colour = "Commercial Ratio Long")) +
# scale_color_manual(values = "green") +
# #labs(colour = "Commercials", x="Date", y="Ratio")
# xlab("sdfs") + ylab("asdfsadf")
xlab("p2x") + ylab("p2y")
#plot(p2)
p2 <- ggplotly(p2)
#p2
p3 <- ggplot(df_max) +
geom_line(aes(x = date_temp, y = ratio2_temp, colour = "Commercial Ratio Short")) +
# geom_point(aes(x = date_temp, y = ratio2_temp, colour = "Commercial Ratio Short")) +
# scale_color_manual(values = "blue") +
# labs(colour = "Commercials", x="Date", y="Ratio")
xlab("p3x") + ylab("p3y")
#plot(p3)
p3 <- ggplotly(p3) #%>%
#layout(xaxis = list(title = "Date"), yaxis = list(title = "Contracts"))
#p3
# combine on one chart with shared X axis
# fig_temp <- subplot(p1, p2, p3, nrows = 3, shareX = TRUE) %>% layout(hovermode = "x unified", hoversubplots = "axis", titleY = TRUE, titleX = TRUE, margin = 0.05)
fig_temp <- subplot(p1, p2, p3, nrows = 3, titleY = TRUE, shareX = TRUE, margin = 0.05) %>% layout(hovermode = "x unified", hoversubplots = "axis")
annotations = list(
list(
x = 0.2,
y = 1.0,
text = "Plot 1",
xref = "paper",
yref = "paper",
xanchor = "center",
yanchor = "bottom",
showarrow = FALSE
),
list(
x = 0.8,
y = 1,
text = "Plot 2",
xref = "paper",
yref = "paper",
xanchor = "center",
yanchor = "bottom",
showarrow = FALSE
),
list(
x = 0.2,
y = 0.4,
text = "Plot 3",
xref = "paper",
yref = "paper",
xanchor = "center",
yanchor = "bottom",
showarrow = FALSE
))
#fig_temp <- fig_temp %>%layout(annotations = annotations)
#fig
fig_temp## Warning: 'layout' objects don't have these attributes: 'hoversubplots'
## Valid attributes include:
## '_deprecated', 'activeshape', 'annotations', 'autosize', 'autotypenumbers', 'calendar', 'clickmode', 'coloraxis', 'colorscale', 'colorway', 'computed', 'datarevision', 'dragmode', 'editrevision', 'editType', 'font', 'geo', 'grid', 'height', 'hidesources', 'hoverdistance', 'hoverlabel', 'hovermode', 'images', 'legend', 'mapbox', 'margin', 'meta', 'metasrc', 'modebar', 'newshape', 'paper_bgcolor', 'plot_bgcolor', 'polar', 'scene', 'selectdirection', 'selectionrevision', 'separators', 'shapes', 'showlegend', 'sliders', 'smith', 'spikedistance', 'template', 'ternary', 'title', 'transition', 'uirevision', 'uniformtext', 'updatemenus', 'width', 'xaxis', 'yaxis', 'barmode', 'bargap', 'mapType'Ok, this might work.
Need to figure out rangeslider too.
library(gridExtra)
# Make three dummy tibbles
df1 <- data.frame(date_temp = c(3, 4, 5), comm_long_temp = c(1000, 1100, 1200))
df1 <- as_tibble(df1)
df2 <- data.frame(date_temp = c(2, 3, 4), ratio1_temp = c(100, 110, 120))
df2 <- as_tibble(df2)
df3 <- data.frame(date_temp = c(0, 1, 2), ratio2_temp = c(10, 11, 12))
df3 <- as_tibble(df3)
# join into one tibble
df_max <- full_join(df1, df2, by = "date_temp")
df_max <- full_join(df_max, df3, by = "date_temp")
# make 3 plotly plots
p1 <- ggplot(df_max) +
geom_line(aes(x = date_temp, y = comm_long_temp, colour = "Commercial Long COT")) +
# geom_point(aes(x = date_temp, y = comm_long_temp, colour = "Commercial Long COT")) +
# scale_color_manual(values = "red") +
# labs(colour = "Commercials", x="Date", y="Ratio")
xlab("p1x") + ylab("p1y")
#plot(p1)
p1 <- ggplotly(p1)
#p1
p2 <- ggplot(df_max) +
geom_line(aes(x = date_temp, y = ratio1_temp, colour = "Commercial Ratio Long")) +
# geom_point(aes(x = date_temp, y = ratio1_temp, colour = "Commercial Ratio Long")) +
# scale_color_manual(values = "green") +
# #labs(colour = "Commercials", x="Date", y="Ratio")
# xlab("sdfs") + ylab("asdfsadf")
xlab("p2x") + ylab("p2y")
#plot(p2)
p2 <- ggplotly(p2)
#p2
p3 <- ggplot(df_max) +
geom_line(aes(x = date_temp, y = ratio2_temp, colour = "Commercial Ratio Short")) +
# geom_point(aes(x = date_temp, y = ratio2_temp, colour = "Commercial Ratio Short")) +
# scale_color_manual(values = "blue") +
# labs(colour = "Commercials", x="Date", y="Ratio")
xlab("p3x") + ylab("p3y")
#plot(p3)
p3 <- ggplotly(p3) #%>%
#layout(xaxis = list(title = "Date"), yaxis = list(title = "Contracts"))
#p3
# combine on one chart with shared X axis
# fig_temp <- subplot(p1, p2, p3, nrows = 3, shareX = TRUE) %>% layout(hovermode = "x unified", hoversubplots = "axis", titleY = TRUE, titleX = TRUE, margin = 0.05)
fig_temp <- subplot(p1, p2, p3, nrows = 3, titleY = TRUE, shareX = TRUE, margin = 0.05) %>% layout(hovermode = "x unified", hoversubplots = "axis", title = "Palladium", xaxis = list(
rangeslider = list(type = "date")))
annotations = list(
list(
x = 0.2,
y = 1.0,
text = "Plot 1",
xref = "paper",
yref = "paper",
xanchor = "center",
yanchor = "bottom",
showarrow = FALSE
),
list(
x = 0.8,
y = 1,
text = "Plot 2",
xref = "paper",
yref = "paper",
xanchor = "center",
yanchor = "bottom",
showarrow = FALSE
),
list(
x = 0.2,
y = 0.4,
text = "Plot 3",
xref = "paper",
yref = "paper",
xanchor = "center",
yanchor = "bottom",
showarrow = FALSE
))
#fig_temp <- fig_temp %>%layout(annotations = annotations)
#fig
fig_temp## Warning: 'layout' objects don't have these attributes: 'hoversubplots'
## Valid attributes include:
## '_deprecated', 'activeshape', 'annotations', 'autosize', 'autotypenumbers', 'calendar', 'clickmode', 'coloraxis', 'colorscale', 'colorway', 'computed', 'datarevision', 'dragmode', 'editrevision', 'editType', 'font', 'geo', 'grid', 'height', 'hidesources', 'hoverdistance', 'hoverlabel', 'hovermode', 'images', 'legend', 'mapbox', 'margin', 'meta', 'metasrc', 'modebar', 'newshape', 'paper_bgcolor', 'plot_bgcolor', 'polar', 'scene', 'selectdirection', 'selectionrevision', 'separators', 'shapes', 'showlegend', 'sliders', 'smith', 'spikedistance', 'template', 'ternary', 'title', 'transition', 'uirevision', 'uniformtext', 'updatemenus', 'width', 'xaxis', 'yaxis', 'barmode', 'bargap', 'mapType'Now, back with all the colors before:
library(gridExtra)
# Make three dummy tibbles
df1 <- data.frame(date_temp = c(3, 4, 5), comm_long_temp = c(1000, 1100, 1200))
df1 <- as_tibble(df1)
df2 <- data.frame(date_temp = c(2, 3, 4), ratio1_temp = c(100, 110, 120))
df2 <- as_tibble(df2)
df3 <- data.frame(date_temp = c(0, 1, 2), ratio2_temp = c(10, 11, 12))
df3 <- as_tibble(df3)
# join into one tibble
df_max <- full_join(df1, df2, by = "date_temp")
df_max <- full_join(df_max, df3, by = "date_temp")
# make 3 plotly plots
p1 <- ggplot(df_max) +
geom_line(aes(x = date_temp, y = comm_long_temp, colour = "Commercial Long COT")) +
geom_point(aes(x = date_temp, y = comm_long_temp, colour = "Commercial Long COT")) +
scale_color_manual(values = "red") +
labs(colour = "Commercials") +
xlab("p1x") + ylab("p1y")
#plot(p1)
p1 <- ggplotly(p1)
#p1
p2 <- ggplot(df_max) +
geom_line(aes(x = date_temp, y = ratio1_temp, colour = "Commercial Ratio Long")) +
geom_point(aes(x = date_temp, y = ratio1_temp, colour = "Commercial Ratio Long")) +
scale_color_manual(values = "green") +
labs(colour = "Commercials") +
# xlab("sdfs") + ylab("asdfsadf")
xlab("p2x") + ylab("p2y")
#plot(p2)
p2 <- ggplotly(p2)
#p2
p3 <- ggplot(df_max) +
geom_line(aes(x = date_temp, y = ratio2_temp, colour = "Commercial Ratio Short")) +
geom_point(aes(x = date_temp, y = ratio2_temp, colour = "Commercial Ratio Short")) +
scale_color_manual(values = "blue") +
labs(colour = "Commercials") +
xlab("p3x") + ylab("p3y")
#plot(p3)
p3 <- ggplotly(p3) #%>%
#layout(xaxis = list(title = "Date"), yaxis = list(title = "Contracts"))
#p3
# combine on one chart with shared X axis
# fig_temp <- subplot(p1, p2, p3, nrows = 3, shareX = TRUE) %>% layout(hovermode = "x unified", hoversubplots = "axis", titleY = TRUE, titleX = TRUE, margin = 0.05)
fig_temp <- subplot(p1, p2, p3, nrows = 3, titleY = TRUE, shareX = TRUE, margin = 0.05) %>% layout(hovermode = "x unified", hoversubplots = "axis", title = "Palladium", xaxis = list(
rangeslider = list(type = "date")))
annotations = list(
list(
x = 0.3,
y = 1.0,
text = "Plot 1",
xref = "paper",
yref = "paper",
xanchor = "center",
yanchor = "bottom",
showarrow = FALSE
),
list(
x = 0.3,
y = 0.6,
text = "Plot 2",
xref = "paper",
yref = "paper",
xanchor = "center",
yanchor = "bottom",
showarrow = FALSE
),
list(
x = 0.3,
y = 0.3,
text = "Plot 3",
xref = "paper",
yref = "paper",
xanchor = "center",
yanchor = "bottom",
showarrow = FALSE
))
fig_temp <- fig_temp %>%layout(annotations = annotations)
#fig
fig_temp## Warning: 'layout' objects don't have these attributes: 'hoversubplots'
## Valid attributes include:
## '_deprecated', 'activeshape', 'annotations', 'autosize', 'autotypenumbers', 'calendar', 'clickmode', 'coloraxis', 'colorscale', 'colorway', 'computed', 'datarevision', 'dragmode', 'editrevision', 'editType', 'font', 'geo', 'grid', 'height', 'hidesources', 'hoverdistance', 'hoverlabel', 'hovermode', 'images', 'legend', 'mapbox', 'margin', 'meta', 'metasrc', 'modebar', 'newshape', 'paper_bgcolor', 'plot_bgcolor', 'polar', 'scene', 'selectdirection', 'selectionrevision', 'separators', 'shapes', 'showlegend', 'sliders', 'smith', 'spikedistance', 'template', 'ternary', 'title', 'transition', 'uirevision', 'uniformtext', 'updatemenus', 'width', 'xaxis', 'yaxis', 'barmode', 'bargap', 'mapType'Ok, back to Palladium:
https://github.com/plotly/plotly.R/issues/1415
comm_long_short_maxmin <- ggplot(palladium_FUT15_16_temp_2_max_and_min) +
# line for Commercial Long
geom_line(aes(x=date,y=Comm_Positions_Long_All, colour = "Long"), size = 1) +
# line for Commercial Short
geom_line(aes(x=date,y=Comm_Positions_Short_All, colour = "Short"), size = 1) +
# better colors and legend
scale_color_manual(values = c("Short" = "red", "Long" = "darkgreen")) +
labs(colour = "Commercials") +
xlab("Date") + ylab("Contracts")
# add label
# labs(
# y="Commercial Interest",
# x="Date",
# title = "Commercial Futures Interest for Palladium As Reported by cftc.gov"
# )
# Redo long short ratio
comm_long_short_ratio_maxmin <- ggplot(palladium_FUT15_16_temp_2_max_and_min) +
geom_line(aes(x=date,y=long_short_ratio, colour = "Long Short Ratio")) +
scale_color_manual(values = "blue") +
labs(colour = "Commercials") +
xlab("Date") + ylab("Ratio")
# labs(
# y="Ratio",
# x="Date",
# title="Ratio of Commercial Long Interest to Commercial Short Interest - Palladium Futures"
# )
# Redo short long ratio
comm_short_long_ratio_maxmin <- ggplot(palladium_FUT15_16_temp_2_max_and_min) +
geom_line(aes(x=date,y=short_long_ratio, colour = "Short Long Ratio")) +
scale_color_manual(values = "purple") +
labs(colour = "Commercials") +
xlab("Date") + ylab("Ratio")
# labs(
# y="Ratio",
# x="Date",
# title="Ratio of Commercial Short Interest to Commercial Long Interest - Palladium Futures"
# )
palladium_price_maxmin <- ggplot(palladium_FUT15_16_temp_2_max_and_min) +
geom_line(aes(x=date,y=adjusted, colour = "Price")) +
scale_color_manual(values = "orange") +
labs(colour = "Commercials") +
xlab("Date") + ylab("Dollars")
# labs(
# y="Ratio",
# x="Date",
# title="Price of PALL ETF"
# )
#plot(palladium_price)
max_gain_maxmin <- ggplot(palladium_FUT15_16_temp_2_max_and_min) +
geom_line(aes(x=date,y=max_gain, colour = "Max Gain")) +
scale_color_manual(values = "yellow") +
labs(colour = "Commercials") +
xlab("Date") + ylab("Percent")
max_loss_maxmin <- ggplot(palladium_FUT15_16_temp_2_max_and_min) +
geom_line(aes(x=date,y=max_loss, colour = "Max Loss")) +
scale_color_manual(values = "brown") +
labs(colour = "Commercials") +
xlab("Date") + ylab("Percent")
fig1 <- ggplotly(comm_long_short_maxmin) #%>% layout(yaxis = list(title = "yaxis test tile"))
fig2 <- ggplotly(comm_long_short_ratio_maxmin)
fig3 <- ggplotly(comm_short_long_ratio_maxmin)
fig4 <- ggplotly(palladium_price_maxmin)
fig5 <- ggplotly(max_gain_maxmin)
fig6 <- ggplotly(max_loss_maxmin)
figs_1_2_3_4_5_5_maxmin <- subplot(fig1, fig2, fig3, fig4, fig5, fig6, nrows = 6, titleY = TRUE, shareX = TRUE, margin = 0.05) %>% layout(hovermode = "x unified", hoversubplots = "axis", title = "Palladium", xaxis = list(
rangeslider = list(type = "date")))
figs_1_2_3_4_5_5_maxmin## Warning: 'layout' objects don't have these attributes: 'hoversubplots'
## Valid attributes include:
## '_deprecated', 'activeshape', 'annotations', 'autosize', 'autotypenumbers', 'calendar', 'clickmode', 'coloraxis', 'colorscale', 'colorway', 'computed', 'datarevision', 'dragmode', 'editrevision', 'editType', 'font', 'geo', 'grid', 'height', 'hidesources', 'hoverdistance', 'hoverlabel', 'hovermode', 'images', 'legend', 'mapbox', 'margin', 'meta', 'metasrc', 'modebar', 'newshape', 'paper_bgcolor', 'plot_bgcolor', 'polar', 'scene', 'selectdirection', 'selectionrevision', 'separators', 'shapes', 'showlegend', 'sliders', 'smith', 'spikedistance', 'template', 'ternary', 'title', 'transition', 'uirevision', 'uniformtext', 'updatemenus', 'width', 'xaxis', 'yaxis', 'barmode', 'bargap', 'mapType'now make bigger?
comm_long_short_maxmin <- ggplot(palladium_FUT15_16_temp_2_max_and_min) +
# line for Commercial Long
geom_line(aes(x=date,y=Comm_Positions_Long_All, colour = "Long"), size = 1) +
# line for Commercial Short
geom_line(aes(x=date,y=Comm_Positions_Short_All, colour = "Short"), size = 1) +
# better colors and legend
scale_color_manual(values = c("Short" = "red", "Long" = "darkgreen")) +
labs(colour = "Commercials") +
xlab("Date") + ylab("Contracts")
# add label
# labs(
# y="Commercial Interest",
# x="Date",
# title = "Commercial Futures Interest for Palladium As Reported by cftc.gov"
# )
# Redo long short ratio
comm_long_short_ratio_maxmin <- ggplot(palladium_FUT15_16_temp_2_max_and_min) +
geom_line(aes(x=date,y=long_short_ratio, colour = "Long Short Ratio")) +
scale_color_manual(values = "blue") +
labs(colour = "Commercials") +
xlab("Date") + ylab("Ratio")
# labs(
# y="Ratio",
# x="Date",
# title="Ratio of Commercial Long Interest to Commercial Short Interest - Palladium Futures"
# )
# Redo short long ratio
comm_short_long_ratio_maxmin <- ggplot(palladium_FUT15_16_temp_2_max_and_min) +
geom_line(aes(x=date,y=short_long_ratio, colour = "Short Long Ratio")) +
scale_color_manual(values = "purple") +
labs(colour = "Commercials") +
xlab("Date") + ylab("Ratio")
# labs(
# y="Ratio",
# x="Date",
# title="Ratio of Commercial Short Interest to Commercial Long Interest - Palladium Futures"
# )
palladium_price_maxmin <- ggplot(palladium_FUT15_16_temp_2_max_and_min) +
geom_line(aes(x=date,y=adjusted, colour = "Price")) +
scale_color_manual(values = "orange") +
labs(colour = "Commercials") +
xlab("Date") + ylab("Dollars")
# labs(
# y="Ratio",
# x="Date",
# title="Price of PALL ETF"
# )
#plot(palladium_price)
max_gain_maxmin <- ggplot(palladium_FUT15_16_temp_2_max_and_min) +
geom_line(aes(x=date,y=max_gain, colour = "Max Gain")) +
scale_color_manual(values = "yellow") +
labs(colour = "Commercials") +
xlab("Date") + ylab("Percent")
max_loss_maxmin <- ggplot(palladium_FUT15_16_temp_2_max_and_min) +
geom_line(aes(x=date,y=max_loss, colour = "Max Loss")) +
scale_color_manual(values = "brown") +
labs(colour = "Commercials") +
xlab("Date") + ylab("Percent")
fig1 <- ggplotly(comm_long_short_maxmin) #%>% layout(yaxis = list(title = "yaxis test tile"))
fig2 <- ggplotly(comm_long_short_ratio_maxmin)
fig3 <- ggplotly(comm_short_long_ratio_maxmin)
fig4 <- ggplotly(palladium_price_maxmin)
fig5 <- ggplotly(max_gain_maxmin)
fig6 <- ggplotly(max_loss_maxmin)
# remove margin for less space in between
figs_1_2_3_4_5_5_maxmin <- subplot(fig1, fig2, fig3, fig4, fig5, fig6, nrows = 6, titleY = TRUE, shareX = TRUE) %>% layout(hovermode = "x unified", hoversubplots = "axis", title = "Palladium", xaxis = list(
rangeslider = list(type = "date")), autosize = F, width = 1000, height = 1000)## Warning: Specifying width/height in layout() is now deprecated.
## Please specify in ggplotly() or plot_ly()figs_1_2_3_4_5_5_maxmin## Warning: 'layout' objects don't have these attributes: 'hoversubplots'
## Valid attributes include:
## '_deprecated', 'activeshape', 'annotations', 'autosize', 'autotypenumbers', 'calendar', 'clickmode', 'coloraxis', 'colorscale', 'colorway', 'computed', 'datarevision', 'dragmode', 'editrevision', 'editType', 'font', 'geo', 'grid', 'height', 'hidesources', 'hoverdistance', 'hoverlabel', 'hovermode', 'images', 'legend', 'mapbox', 'margin', 'meta', 'metasrc', 'modebar', 'newshape', 'paper_bgcolor', 'plot_bgcolor', 'polar', 'scene', 'selectdirection', 'selectionrevision', 'separators', 'shapes', 'showlegend', 'sliders', 'smith', 'spikedistance', 'template', 'ternary', 'title', 'transition', 'uirevision', 'uniformtext', 'updatemenus', 'width', 'xaxis', 'yaxis', 'barmode', 'bargap', 'mapType'
## now add range buttons for last xx months, xx years, all time
https://plotly.com/r/range-slider/
https://stackoverflow.com/questions/59877688/r-plotly-rangeselector-autoselect-1w
Might now work with ggplot2 and plotly https://community.plotly.com/t/rangeselector-buttons-with-plot-ly-vs-ggplotly/1562
can’t make it work yet
but maybe dynamic tricks???
https://github.com/plotly/plotly.R/issues/646
comm_long_short_maxmin <- ggplot(palladium_FUT15_16_temp_2_max_and_min) +
# line for Commercial Long
geom_line(aes(x=date,y=Comm_Positions_Long_All, colour = "Long"), size = 1) +
# line for Commercial Short
geom_line(aes(x=date,y=Comm_Positions_Short_All, colour = "Short"), size = 1) +
# better colors and legend
scale_color_manual(values = c("Short" = "red", "Long" = "darkgreen")) +
labs(colour = "Commercials") +
xlab("Date") + ylab("Contracts")
# Redo long short ratio
comm_long_short_ratio_maxmin <- ggplot(palladium_FUT15_16_temp_2_max_and_min) +
geom_line(aes(x=date,y=long_short_ratio, colour = "Long Short Ratio")) +
scale_color_manual(values = "blue") +
labs(colour = "Commercials") +
xlab("Date") + ylab("Ratio")
# Redo short long ratio
comm_short_long_ratio_maxmin <- ggplot(palladium_FUT15_16_temp_2_max_and_min) +
geom_line(aes(x=date,y=short_long_ratio, colour = "Short Long Ratio")) +
scale_color_manual(values = "purple") +
labs(colour = "Commercials") +
xlab("Date") + ylab("Ratio")
palladium_price_maxmin <- ggplot(palladium_FUT15_16_temp_2_max_and_min) +
geom_line(aes(x=date,y=adjusted, colour = "Price")) +
scale_color_manual(values = "orange") +
labs(colour = "Commercials") +
xlab("Date") + ylab("Dollars")
max_gain_maxmin <- ggplot(palladium_FUT15_16_temp_2_max_and_min) +
geom_line(aes(x=date,y=max_gain, colour = "Max Gain")) +
scale_color_manual(values = "magenta") +
labs(colour = "Commercials") +
xlab("Date") + ylab("Percent")
max_loss_maxmin <- ggplot(palladium_FUT15_16_temp_2_max_and_min) +
geom_line(aes(x=date,y=max_loss, colour = "Max Loss")) +
scale_color_manual(values = "brown") +
labs(colour = "Commercials") +
xlab("Date") + ylab("Percent")
fig1 <- ggplotly(comm_long_short_maxmin) #%>% layout(yaxis = list(title = "yaxis test tile"))
fig2 <- ggplotly(comm_long_short_ratio_maxmin)
fig3 <- ggplotly(comm_short_long_ratio_maxmin)
fig4 <- ggplotly(palladium_price_maxmin)
fig5 <- ggplotly(max_gain_maxmin)
fig6 <- ggplotly(max_loss_maxmin)
# remove margin for less space in between
figs_1_2_3_4_5_5_maxmin <- subplot(fig1, fig2, fig3, fig4, fig5, fig6, nrows = 6, titleY = TRUE, shareX = TRUE) %>%
layout(hovermode = "x unified", hoversubplots = "axis", title = "Palladium",
xaxis = list(
rangeselector = list(
buttons = list(
list(
count = 3,
label = "3 mo",
step = "month",
stepmode = "backward"),
list(
count = 6,
label = "6 mo",
step = "month",
stepmode = "backward"),
list(
count = 1,
label = "1 yr",
step = "year",
stepmode = "backward"),
list(
count = 1,
label = "YTD",
step = "year",
stepmode = "todate"),
list(step = "all"))),
rangeslider = list(type = "date")), autosize = F, width = 1000, height = 1000)## Warning: Specifying width/height in layout() is now deprecated.
## Please specify in ggplotly() or plot_ly()figs_1_2_3_4_5_5_maxmin## Warning: 'layout' objects don't have these attributes: 'hoversubplots'
## Valid attributes include:
## '_deprecated', 'activeshape', 'annotations', 'autosize', 'autotypenumbers', 'calendar', 'clickmode', 'coloraxis', 'colorscale', 'colorway', 'computed', 'datarevision', 'dragmode', 'editrevision', 'editType', 'font', 'geo', 'grid', 'height', 'hidesources', 'hoverdistance', 'hoverlabel', 'hovermode', 'images', 'legend', 'mapbox', 'margin', 'meta', 'metasrc', 'modebar', 'newshape', 'paper_bgcolor', 'plot_bgcolor', 'polar', 'scene', 'selectdirection', 'selectionrevision', 'separators', 'shapes', 'showlegend', 'sliders', 'smith', 'spikedistance', 'template', 'ternary', 'title', 'transition', 'uirevision', 'uniformtext', 'updatemenus', 'width', 'xaxis', 'yaxis', 'barmode', 'bargap', 'mapType'
now add all years of palladium data
# GET ALL DATA FROM ALL YEARS FOR PALL
# work with Palladium
# Use .csv files
palladium_FUT86_06 <- read_xls("FUT86_06.xls")
#palladium_FUT86_06
palladium_FUT86_06 <- palladium_FUT86_06 %>% filter(str_detect( Market_and_Exchange_Names, "PALL"))
#palladium_FUT86_06
# second column "As_of...." is in character form. Need number to rbind with others
palladium_FUT86_06 <- palladium_FUT86_06 %>% mutate(As_of_Date_In_Form_YYMMDD = parse_number(As_of_Date_In_Form_YYMMDD))
#palladium_FUT86_06
palladium_FUT07_14 <- read_xls("FUT07_14.xls")
#palladium_FUT07_14
palladium_FUT07_14 <- palladium_FUT07_14 %>% filter(str_detect( Market_and_Exchange_Names, "PALL"))
# second column "As_of...." is in character form. Need number to rbind with others
palladium_FUT07_14 <- palladium_FUT07_14 %>% mutate(As_of_Date_In_Form_YYMMDD = parse_number(As_of_Date_In_Form_YYMMDD))
#palladium_FUT07_14
palladium_FUT15_16 <- read_xls("FUT15_16.xls")
#palladium_FUT15_16
palladium_FUT15_16 <- palladium_FUT15_16 %>% filter(str_detect( Market_and_Exchange_Names, "PALL"))
#palladium_FUT15_16
# second column "As_of...." is in character form. Need number to rbind with others
palladium_FUT15_16 <- palladium_FUT15_16 %>% mutate(As_of_Date_In_Form_YYMMDD = parse_number(As_of_Date_In_Form_YYMMDD))
#palladium_FUT15_16
palladium_FUT17 <- read_xls("FUT17.xls")
#palladium_FUT17
palladium_FUT17 <- palladium_FUT17 %>% filter(str_detect( Market_and_Exchange_Names, "PALL"))
palladium_FUT18 <- read_xls("FUT18.xls")
#palladium_FUT18
palladium_FUT18 <- palladium_FUT18 %>% filter(str_detect( Market_and_Exchange_Names, "PALL"))
palladium_FUT19 <- read_xls("FUT19.xls")
#palladium_FUT19
palladium_FUT19 <- palladium_FUT19 %>% filter(str_detect( Market_and_Exchange_Names, "PALL"))
palladium_FUT20 <- read_xls("FUT20.xls")
#palladium_FUT20
palladium_FUT20 <- palladium_FUT20 %>% filter(str_detect( Market_and_Exchange_Names, "PALL"))
palladium_FUT21 <- read_xls("FUT21.xls")
#palladium_FUT21
palladium_FUT21 <- palladium_FUT21 %>% filter(str_detect( Market_and_Exchange_Names, "PALL"))
palladium_FUT22 <- read_xls("FUT22.xls")
#palladium_FUT22
palladium_FUT22 <- palladium_FUT22 %>% filter(str_detect( Market_and_Exchange_Names, "PALL"))
palladium_FUT23 <- read_xls("FUT23.xls")
#palladium_FUT23
palladium_FUT23 <- palladium_FUT23 %>% filter(str_detect( Market_and_Exchange_Names, "PALL"))
palladium_FUT24 <- read_xls("FUT24.xls")
#palladium_FUT24
palladium_FUT24 <- palladium_FUT24 %>% filter(str_detect( Market_and_Exchange_Names, "PALL"))
palladium_all <- bind_rows(palladium_FUT24, palladium_FUT23, palladium_FUT22, palladium_FUT21, palladium_FUT20, palladium_FUT19, palladium_FUT18, palladium_FUT17, palladium_FUT15_16, palladium_FUT07_14, palladium_FUT86_06)
# one file has double and one has character for date
# fixed above
palladium_all <- palladium_all %>% mutate(long_short_ratio = Comm_Positions_Long_All/Comm_Positions_Short_All)
palladium_all <- palladium_all %>% mutate(short_long_ratio = Comm_Positions_Short_All/Comm_Positions_Long_All)
# get all PALL prices
# tq_get from tidyquant returns a tibble
# PALL_price <- tq_get('PALL', from = '2000-01-01', to = "2025-01-01", get = "stock.prices" )
# head(PALL_price)
# PALL doesn't start until the first week of 2010
# leaving out the end date gets up to current data
# leaving out the start date only begins in 2014 for some reason
# look into this
PALL_price <- tq_get('PALL', from = '2000-01-01', get = "stock.prices")
head(PALL_price)## # A tibble: 6 × 8
## symbol date open high low close volume adjusted
## <chr> <date> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
## 1 PALL 2010-01-08 43.7 45 43.2 43.2 294900 43.2
## 2 PALL 2010-01-11 43.7 43.7 43.2 43.7 338800 43.7
## 3 PALL 2010-01-12 43.6 43.8 42.3 42.3 340900 42.3
## 4 PALL 2010-01-13 42.4 42.7 41.7 42.7 1011400 42.7
## 5 PALL 2010-01-14 43.9 45.0 43.9 44.8 364600 44.8
## 6 PALL 2010-01-15 45.2 45.8 44.4 45.8 442300 45.8tail(PALL_price)## # A tibble: 6 × 8
## symbol date open high low close volume adjusted
## <chr> <date> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
## 1 PALL 2024-09-27 94.7 96.0 92.8 92.9 131300 92.9
## 2 PALL 2024-09-30 91.5 92.2 90.1 91.7 112300 91.7
## 3 PALL 2024-10-01 91.7 92.9 90.6 92.2 69500 92.2
## 4 PALL 2024-10-02 94.1 94.1 92.6 93.9 82200 93.9
## 5 PALL 2024-10-03 90.9 92.3 90.4 92.2 78400 92.2
## 6 PALL 2024-10-04 92.4 93.3 91.4 93 197000 93# need to combine PALL_price and palladium_all
library(lubridate)
# rename column in palladium_FUT15_15 to match pall_tq
palladium_all <- rename(palladium_all, date=Report_Date_as_MM_DD_YYYY)
# make sure in ymd format
palladium_all$date <- ymd(palladium_all$date)
# find where join might work
#palladium_FUT15_16_with_price <-
anti_join(palladium_all, PALL_price, by = "date")## # A tibble: 990 × 128
## Market_and_Exchange_Names As_of_Date_In_Form_YYMMDD date
## <chr> <dbl> <date>
## 1 PALLADIUM - NEW YORK MERCANTILE EXCHANGE 121030 2012-10-30
## 2 PALLADIUM - NEW YORK MERCANTILE EXCHANGE 100105 2010-01-05
## 3 PALLADIUM - NEW YORK MERCANTILE EXCHANGE 91229 2009-12-29
## 4 PALLADIUM - NEW YORK MERCANTILE EXCHANGE 91222 2009-12-22
## 5 PALLADIUM - NEW YORK MERCANTILE EXCHANGE 91215 2009-12-15
## 6 PALLADIUM - NEW YORK MERCANTILE EXCHANGE 91208 2009-12-08
## 7 PALLADIUM - NEW YORK MERCANTILE EXCHANGE 91201 2009-12-01
## 8 PALLADIUM - NEW YORK MERCANTILE EXCHANGE 91124 2009-11-24
## 9 PALLADIUM - NEW YORK MERCANTILE EXCHANGE 91117 2009-11-17
## 10 PALLADIUM - NEW YORK MERCANTILE EXCHANGE 91109 2009-11-09
## # ℹ 980 more rows
## # ℹ 125 more variables: CFTC_Contract_Market_Code <chr>,
## # CFTC_Market_Code <chr>, CFTC_Region_Code <chr>, CFTC_Commodity_Code <chr>,
## # Open_Interest_All <dbl>, NonComm_Positions_Long_All <dbl>,
## # NonComm_Positions_Short_All <dbl>, NonComm_Postions_Spread_All <dbl>,
## # Comm_Positions_Long_All <dbl>, Comm_Positions_Short_All <dbl>,
## # Tot_Rept_Positions_Long_All <dbl>, Tot_Rept_Positions_Short_All <dbl>, …semi_join(palladium_all, PALL_price, by = "date")## # A tibble: 763 × 128
## Market_and_Exchange_Names As_of_Date_In_Form_YYMMDD date
## <chr> <dbl> <date>
## 1 PALLADIUM - NEW YORK MERCANTILE EXCHANGE 240827 2024-08-27
## 2 PALLADIUM - NEW YORK MERCANTILE EXCHANGE 240820 2024-08-20
## 3 PALLADIUM - NEW YORK MERCANTILE EXCHANGE 240813 2024-08-13
## 4 PALLADIUM - NEW YORK MERCANTILE EXCHANGE 240806 2024-08-06
## 5 PALLADIUM - NEW YORK MERCANTILE EXCHANGE 240730 2024-07-30
## 6 PALLADIUM - NEW YORK MERCANTILE EXCHANGE 240723 2024-07-23
## 7 PALLADIUM - NEW YORK MERCANTILE EXCHANGE 240716 2024-07-16
## 8 PALLADIUM - NEW YORK MERCANTILE EXCHANGE 240709 2024-07-09
## 9 PALLADIUM - NEW YORK MERCANTILE EXCHANGE 240702 2024-07-02
## 10 PALLADIUM - NEW YORK MERCANTILE EXCHANGE 240625 2024-06-25
## # ℹ 753 more rows
## # ℹ 125 more variables: CFTC_Contract_Market_Code <chr>,
## # CFTC_Market_Code <chr>, CFTC_Region_Code <chr>, CFTC_Commodity_Code <chr>,
## # Open_Interest_All <dbl>, NonComm_Positions_Long_All <dbl>,
## # NonComm_Positions_Short_All <dbl>, NonComm_Postions_Spread_All <dbl>,
## # Comm_Positions_Long_All <dbl>, Comm_Positions_Short_All <dbl>,
## # Tot_Rept_Positions_Long_All <dbl>, Tot_Rept_Positions_Short_All <dbl>, …palladium_all <- left_join(palladium_all, PALL_price, by = "date")
head(palladium_all$adjusted)## [1] 89.20 85.00 85.90 80.34 82.00 85.31tail(palladium_all$adjusted)## [1] NA NA NA NA NA NA# now add max and min
library(slider)
# for now using without pivot
palladium_all <- palladium_all %>%
mutate(six_month_max = slide_dbl(adjusted, max, .before = 26, .complete = TRUE)) %>%
mutate(six_month_min = slide_dbl(adjusted, min, .before = 26, .complete = TRUE)) %>%
mutate(max_gain = 100*(six_month_max-adjusted)/adjusted) %>%
mutate(max_loss = 100*(six_month_min-adjusted)/adjusted)
head(palladium_all,20)[c("adjusted","six_month_max","six_month_min","max_gain","max_loss")]## # A tibble: 20 × 5
## adjusted six_month_max six_month_min max_gain max_loss
## <dbl> <dbl> <dbl> <dbl> <dbl>
## 1 89.2 NA NA NA NA
## 2 85 NA NA NA NA
## 3 85.9 NA NA NA NA
## 4 80.3 NA NA NA NA
## 5 82 NA NA NA NA
## 6 85.3 NA NA NA NA
## 7 88.5 NA NA NA NA
## 8 90.0 NA NA NA NA
## 9 93.3 NA NA NA NA
## 10 86.7 NA NA NA NA
## 11 81.8 NA NA NA NA
## 12 81.7 NA NA NA NA
## 13 84.1 NA NA NA NA
## 14 90.2 NA NA NA NA
## 15 94.8 NA NA NA NA
## 16 90.0 NA NA NA NA
## 17 89.3 NA NA NA NA
## 18 87.8 NA NA NA NA
## 19 93.7 NA NA NA NA
## 20 93.3 NA NA NA NAtail(palladium_all,20)[c("adjusted","six_month_max","six_month_min","max_gain","max_loss")]## # A tibble: 20 × 5
## adjusted six_month_max six_month_min max_gain max_loss
## <dbl> <dbl> <dbl> <dbl> <dbl>
## 1 NA NA NA NA NA
## 2 NA NA NA NA NA
## 3 NA NA NA NA NA
## 4 NA NA NA NA NA
## 5 NA NA NA NA NA
## 6 NA NA NA NA NA
## 7 NA NA NA NA NA
## 8 NA NA NA NA NA
## 9 NA NA NA NA NA
## 10 NA NA NA NA NA
## 11 NA NA NA NA NA
## 12 NA NA NA NA NA
## 13 NA NA NA NA NA
## 14 NA NA NA NA NA
## 15 NA NA NA NA NA
## 16 NA NA NA NA NA
## 17 NA NA NA NA NA
## 18 NA NA NA NA NA
## 19 NA NA NA NA NA
## 20 NA NA NA NA NAcomm_long_short_maxmin <- ggplot(palladium_all) +
# line for Commercial Long
geom_line(aes(x=date,y=Comm_Positions_Long_All, colour = "Long"), size = 1) +
# line for Commercial Short
geom_line(aes(x=date,y=Comm_Positions_Short_All*-1, colour = "Short"), size = 1) +
# better colors and legend
scale_color_manual(values = c("Short" = "red", "Long" = "darkgreen")) +
labs(colour = "Commercials") +
xlab("Date") + ylab("Contracts")
# add label
# labs(
# y="Commercial Interest",
# x="Date",
# title = "Commercial Futures Interest for Palladium As Reported by cftc.gov"
# )
# Redo long short ratio
comm_long_short_ratio_maxmin <- ggplot(palladium_all) +
geom_line(aes(x=date,y=long_short_ratio, colour = "Long Short Ratio")) +
scale_color_manual(values = "blue") +
labs(colour = "Commercials") +
xlab("Date") + ylab("Ratio")
# labs(
# y="Ratio",
# x="Date",
# title="Ratio of Commercial Long Interest to Commercial Short Interest - Palladium Futures"
# )
# Redo short long ratio
comm_short_long_ratio_maxmin <- ggplot(palladium_all) +
geom_line(aes(x=date,y=short_long_ratio, colour = "Short Long Ratio")) +
scale_color_manual(values = "purple") +
labs(colour = "Commercials") +
xlab("Date") + ylab("Ratio")
# labs(
# y="Ratio",
# x="Date",
# title="Ratio of Commercial Short Interest to Commercial Long Interest - Palladium Futures"
# )
palladium_price_maxmin <- ggplot(palladium_all) +
geom_line(aes(x=date,y=adjusted, colour = "Price - PALL ETF")) +
scale_color_manual(values = "orange") +
labs(colour = "Commercials") +
xlab("Date") + ylab("Dollars")
# labs(
# y="Ratio",
# x="Date",
# title="Price of PALL ETF"
# )
#plot(palladium_price)
max_gain_maxmin <- ggplot(palladium_all) +
geom_line(aes(x=date,y=max_gain, colour = "Max Gain - 6 months")) +
scale_color_manual(values = "magenta") +
labs(colour = "Commercials") +
xlab("Date") + ylab("Percent")
max_loss_maxmin <- ggplot(palladium_all) +
geom_line(aes(x=date,y=max_loss, colour = "Max Loss - 6 months")) +
scale_color_manual(values = "brown") +
labs(colour = "Commercials") +
xlab("Date") + ylab("Percent")
fig1 <- ggplotly(comm_long_short_maxmin) #%>% layout(yaxis = list(title = "yaxis test tile"))
fig2 <- ggplotly(comm_long_short_ratio_maxmin)
fig3 <- ggplotly(comm_short_long_ratio_maxmin)
fig4 <- ggplotly(palladium_price_maxmin)
fig5 <- ggplotly(max_gain_maxmin)
fig6 <- ggplotly(max_loss_maxmin)
# remove margin for less space in between
figs_1_2_3_4_5_5_maxmin <- subplot(fig1, fig2, fig3, fig4, fig5, fig6, nrows = 6, titleY = TRUE, shareX = TRUE) %>% layout(hovermode = "x unified", hoversubplots = "axis", title = "Palladium", xaxis = list(
rangeslider = list(type = "date")), autosize = F, width = 1000, height = 1000)## Warning: Specifying width/height in layout() is now deprecated.
## Please specify in ggplotly() or plot_ly()## Warning: Specifying width/height in layout() is now deprecated.
## Please specify in ggplotly() or plot_ly()
figs_1_2_3_4_5_5_maxmin## Warning: 'layout' objects don't have these attributes: 'hoversubplots'
## Valid attributes include:
## '_deprecated', 'activeshape', 'annotations', 'autosize', 'autotypenumbers', 'calendar', 'clickmode', 'coloraxis', 'colorscale', 'colorway', 'computed', 'datarevision', 'dragmode', 'editrevision', 'editType', 'font', 'geo', 'grid', 'height', 'hidesources', 'hoverdistance', 'hoverlabel', 'hovermode', 'images', 'legend', 'mapbox', 'margin', 'meta', 'metasrc', 'modebar', 'newshape', 'paper_bgcolor', 'plot_bgcolor', 'polar', 'scene', 'selectdirection', 'selectionrevision', 'separators', 'shapes', 'showlegend', 'sliders', 'smith', 'spikedistance', 'template', 'ternary', 'title', 'transition', 'uirevision', 'uniformtext', 'updatemenus', 'width', 'xaxis', 'yaxis', 'barmode', 'bargap', 'mapType'
add bar plot with long and short against each other
add 12 months max gain and loss
add amount of interest across all
add non-commercials
add percentile for long and short interest and ratios
plotly may be able to do dual axes
https://plotly.com/r/multiple-axes/ Plotly stuff:
book
https://community.plotly.com/t/announcing-plotly-r-4-9-2-9000-shape-drawing-unified-hover-and-rangebreaks/40086
https://plotly.com/r/subplots/
https://plotly.com/r/reference/layout/#layout-hovermode
https://plotly.com/r/hover-text-and-formatting/
https://stackoverflow.com/questions/58103196/how-to-facet-a-plot-ly-chart
https://stackoverflow.com/questions/54561931/r-plotly-link-subplots-so-that-multiple-tooltips-shown-on-hover
https://stackoverflow.com/questions/69903147/how-to-link-traces-between-subplots-in-r-plotly-with-shared-y-axis-so-that-hover
https://stackoverflow.com/questions/66767067/how-to-change-order-of-r-plotly-hover-labels-with-x-unified-mode
https://plotly.com/r/time-series/#hover-templates-with-mixtures-of-period-data
https://www.youtube.com/watch?v=72D6bepQWXQ
https://stackoverflow.com/questions/54561931/r-plotly-link-subplots-so-that-multiple-tooltips-shown-on-hover
https://plotly.com/r/graphing-multiple-chart-types/