CatFish

II) My second dataset is from Chad Smith based on ‘catfish’ survey.I calculated the yearwise and size category wise numbers.Starting with crunching data and tidying it.

Link to dataset: CatfishFarm.xlsx

library(tidyr)
library(dplyr)

## 
## Attaching package: 'dplyr'

## The following objects are masked from 'package:stats':
## 
##     filter, lag

## The following objects are masked from 'package:base':
## 
##     intersect, setdiff, setequal, union

library(ggplot2)
library(reshape2)

## 
## Attaching package: 'reshape2'

## The following object is masked from 'package:tidyr':
## 
##     smiths

CFData <- read.csv("CatfishFarm.csv")
colnames(CFData) = as.character(unlist(CFData[1,]))
CFData = CFData[-1,]
CFData = CFData[-7:-8,]
CFData

##        Size category    1992    1993    1994    1995    1996    1997
## 2       Broodfish 1/   1,491   1,169   1,183   1,301   1,171   1,163
## 3  Fingerling/fry 2/ 849,412 669,491 648,628 724,693 823,397 873,457
## 4        Stockers 3/ 634,353 571,254 548,207 554,342 627,834 754,816
## 5  Small foodsize 4/ 166,731 153,600 134,314 138,160 156,297 178,448
## 6 Medium foodsize 5/  70,495  61,894  48,851  59,159  64,858  84,725
## 7  Large foodsize 6/   6,769   6,698   5,196   4,536   6,644   7,810
##      1998    1999      2000      2001      2002    2003    2004    2005
## 2   1,187   1,155     1,377     1,327     1,171   1,303   1,113   1,053
## 3 975,542 986,368 1,053,300 1,023,533 1,066,400 990,163 745,849 712,144
## 4 607,878 678,682   790,683   845,287   676,378 775,226 890,275 660,000
## 5 178,511 182,251   200,032   239,655   287,591 254,920 261,323 243,090
## 6  62,140  63,049    77,149    87,926   106,117 127,908 109,120  95,240
## 7   7,295   9,266     5,812     6,872    10,746  11,195  10,947  10,642
##        2006    2007    2008    2009    2010    2011    2012    2013
## 2     1,091     886     801     704     536     495     562     540
## 3 1,045,266 985,620 951,910 728,340 429,590 568,990 451,100 398,510
## 4   781,958 586,320 688,844 586,069 366,090 380,660 463,485 339,260
## 5   214,848 210,340 204,750 193,870 169,030 115,560 112,970 103,520
## 6   103,591 104,080 107,800 105,610  91,790  54,130  64,740  58,015
## 7    10,823   8,986   9,290   9,316   8,570   6,212   3,595   5,155
##      2014    2015    2016
## 2     650     577     520
## 3 420,060 449,510 328,570
## 4 289,080 248,790 204,800
## 5 102,190  96,810 100,850
## 6  50,600  48,220  45,775
## 7   4,500   5,090   3,520

collecting data from ‘wide’ to ‘tall’ format

TCFData <- CFData %>%
                gather(Year,Budget,c(2:26))

## Warning: attributes are not identical across measure variables;
## they will be dropped

TCFData$Budget <- as.numeric(gsub(",","",TCFData$Budget))
TCFData$Year <- as.numeric(TCFData$Year)
head(TCFData)

##        Size category Year Budget
## 1       Broodfish 1/ 1992   1491
## 2  Fingerling/fry 2/ 1992 849412
## 3        Stockers 3/ 1992 634353
## 4  Small foodsize 4/ 1992 166731
## 5 Medium foodsize 5/ 1992  70495
## 6  Large foodsize 6/ 1992   6769

ggplot(TCFData,aes(x = reorder(Year,Budget) , y = Budget,color = `Size category`,group = `Size category`))+ geom_line(linetype = "dotted") + geom_point(shape = 20,size = 4) + labs(x = "Year", y = "Budget")+ coord_flip()

The graph makes it clear that there is much variation in amount of size -“Fingerling/fry” over the years.Th prices for Broodfish and and Large foodsize fishes are constant.

To analyze data yearwise,we choose to use “Spread” function.

TSCFdata <- spread(TCFData,`Size category`,`Budget`)
TSCFdata <- data.frame(TSCFdata)
head(TSCFdata)

##   Year Broodfish.1. Fingerling.fry.2. Large.foodsize.6. Medium.foodsize.5.
## 1 1992         1491            849412              6769              70495
## 2 1993         1169            669491              6698              61894
## 3 1994         1183            648628              5196              48851
## 4 1995         1301            724693              4536              59159
## 5 1996         1171            823397              6644              64858
## 6 1997         1163            873457              7810              84725
##   Small.foodsize.4. Stockers.3.
## 1            166731      634353
## 2            153600      571254
## 3            134314      548207
## 4            138160      554342
## 5            156297      627834
## 6            178448      754816

TSCFdata <- na.omit(TSCFdata)
TSCFdata$Broodfish.1. <- as.numeric(gsub(",","",TSCFdata$Broodfish.1.)) 
TSCFdata$Fingerling.fry.2. <- as.numeric(gsub(",","",TSCFdata$Fingerling.fry.2.))
TSCFdata$Large.foodsize.6. <- as.numeric(gsub(",","",TSCFdata$Large.foodsize.6.))
TSCFdata$Medium.foodsize.5. <- as.numeric(gsub(",","",TSCFdata$Medium.foodsize.5.))
TSCFdata$Small.foodsize.4. <- as.numeric(gsub(",","",TSCFdata$Small.foodsize.4.))
TSCFdata$Stockers.3. <- as.numeric(gsub(",","",TSCFdata$Stockers.3.)) 
TSCFData1 <- mutate(TSCFdata,Total = rowSums(TSCFdata[,2:7]))

ggplot(TSCFData1,aes(x = reorder(Year,Total) , y = Total))+ geom_line(linetype = "dotted") + geom_point(shape = 20,size = 4) + ggtitle("Year wise comparison") + labs(x = "Year", y = "Total Inventory Amount",srt = 45)+ coord_flip()

## geom_path: Each group consists of only one observation. Do you need to
## adjust the group aesthetic?

Obeservation shows that year 2016 resulted into much less inventory amount for total catfish purchase.