ContextBase Logo



Synopsis

This document utilizes the “QuantMod”, and “PerformanceAnalytics”, R packages for Backtesting of SMA Trading Stategies.



Working Directory, and Required Packages 

# Set Working Directory to folder with project files
setwd("C:/Users/Administrator/Dropbox/Programming/PredictiveAnalytics/TimeSeriesAnalysis/StockTrading/AutomatedTradingStrategies")

# Required Packages
# if (!require("xts")) { install.packages("xts"); require("xts") }
# if (!require("zoo")) { install.packages("zoo"); require("zoo") }
# if (!require("quantmod")) { install.packages("quantmod"); require("quantmod") }
# if (!require("PerformanceAnalytics")) { install.packages("PerformanceAnalytics"); require("PerformanceAnalytics") }
# if (!require("TTR")) { install.packages("TTR"); require("TTR") }

library(xts)
library(zoo)
library(quantmod) # data, plotting, quant modelling
library(PerformanceAnalytics) # performance and risk management
library(TTR)
library(knitr) # formatting tables



Variable SMA (Simple Moving Average) Trading Strategy 

# Downloading Stock Ticker Data from Yahoo Finances
# Enter Ticker Symbol and Date Range
SYMBL <- getSymbols("TSLA", auto.assign=F, from="2015-10-18", to="2017-10-18")

# Display first line of Stock Ticker Data
head(SYMBL, 1)
##            TSLA.Open TSLA.High TSLA.Low TSLA.Close TSLA.Volume
## 2015-10-19     226.5    231.15   224.94      228.1     2507900
##            TSLA.Adjusted
## 2015-10-19         228.1
# Extract the closing prices
data <- SYMBL[,4]

# Calculate the SMA
# Set "n=" to the SMA period
sma <- SMA(Cl(data), n=50)

# Calculate MACD
# Set "nsig=" to the SMA period
macd <- MACD(data, nFast=12, nSlow=26, nSig=50, maType=SMA, percent = FALSE)

# Calculate the buy/sell signals
signal <- Lag(ifelse(macd$macd < macd$signal, -1, 1))

# Calculate the returns from the Trading Strategy
returns <- ROC(data)*signal

# Remove not available values from the returns
returns <- na.omit(returns)

# Calculate the Portfolio
portfolio <- exp(cumsum(returns))



Evaluation of Strategy Performance 

# Table of Calendar Returns
kable(table.CalendarReturns(returns), caption = "Calendar Returns")
Calendar Returns
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec TSLA.Close
2016 NA 0.8 1.3 2.8 -0.1 -1.0 1.8 -0.3 1.6 -1.1 0.1 -0.5 5.5
2017 0.5 -1.5 0.1 1.7 -1.7 0.2 3.5 0.8 0.4 -1.5 NA NA 2.6 
# Table of Drawdowns
kable(table.Drawdowns(returns, top=10), caption = "Table of Drawdowns")
Table of Drawdowns
From Trough To Depth Length To Trough Recovery
2017-02-15 2017-10-06 NA -0.3424 171 163 NA
2016-05-18 2016-11-25 2017-01-23 -0.3109 172 134 38
2016-02-11 2016-03-01 2016-04-06 -0.2678 38 13 25
2016-04-07 2016-04-19 2016-05-03 -0.0696 19 9 10
2017-01-25 2017-02-01 2017-02-06 -0.0212 9 6 3
2016-05-06 2016-05-06 2016-05-09 -0.0159 2 1 1
2016-05-13 2016-05-16 2016-05-17 -0.0049 3 2 1
2016-02-09 2016-02-09 2016-02-10 -0.0018 2 1 1
2016-05-11 2016-05-11 2016-05-12 -0.0013 2 1 1
2017-02-07 2017-02-07 2017-02-08 -0.0011 2 1 1 
# Table of Downside Risk
kable(table.DownsideRisk(returns), caption = "Downside Risk Table")
Downside Risk Table
TSLA.Close
Semi Deviation 0.0165
Gain Deviation 0.0155
Loss Deviation 0.0159
Downside Deviation (MAR=210%) 0.0210
Downside Deviation (Rf=0%) 0.0163
Downside Deviation (0%) 0.0163
Maximum Drawdown 0.3424
Historical VaR (95%) -0.0360
Historical ES (95%) -0.0515
Modified VaR (95%) -0.0375
Modified ES (95%) -0.0554 
# Plot the SMA
# Select next two lines of code, and run
plotSMA <- plot(data, main="Returns with SMA Trend Line"); lines(sma, col="red")

# Plot of Buy/Sell/Hold signals
# Select next two lines of code, and run
plotBuySell <- plot(macd$signal, main="Buy/Sell/Hold Signal
                    Plot");lines(signal,col="red")

# Performance Summary Chart
charts.PerformanceSummary(returns, main = "Performance Summary Chart")

# Create Chart Series with MACD
# Set "signal=" to SMA period
chartSeries(data, name = "Strategy Chart Series", theme = chartTheme('white'),
            TA=c(addVo(), addBBands(), addMACD(signal=50)))