Scribd
Upload
17 views

Downloading and Exploring FTS Data R Code

This document discusses analyzing financial time series data of Microsoft's stock prices from 2011 to 2021. It shows how to download, plot and explore the data. Descriptive statistics like mean, variance, skewness are calculated for simple and log returns. Tests for normality, stationarity and serial correlation are performed. ACF and PACF plots are used to check for autocorrelation in returns and prices.

Uploaded by

Bev
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
17 views

Downloading and Exploring FTS Data R Code

This document discusses analyzing financial time series data of Microsoft's stock prices from 2011 to 2021. It shows how to download, plot and explore the data. Descriptive statistics like mean, variance, skewness are calculated for simple and log returns. Tests for normality, stationarity and serial correlation are performed. ACF and PACF plots are used to check for autocorrelation in returns and prices.

Uploaded by

Bev
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 11

Downloading and Exploring Financial Time series Data

Ms. Matabel Odin

9/18/2023

The following R packages can used to download financial data: library(timeSeries)


library(quantmod) or Download directly from Yahoo Finance or any other source and save
as an xls, xlsx or csv file
Lets load some data
MSFT_data=read.csv(file="D:/undergraduate_notes/fts_sta2420/MSFT.csv") #load
data
summary(MSFT_data)

## Date AdjClose
## Length:121 Min. : 20.59
## Class :character 1st Qu.: 35.48
## Mode :character Median : 55.75
## Mean : 91.06
## 3rd Qu.:127.06
## Max. :301.30

The data corresponds to monthly MSFT adjusted closing prices from November 2011 to
August 2021.
EXPLORATORY DATA ANALYSIS.
Load some packages for investigating properties of financial data
library(fBasics)

## Warning: package 'fBasics' was built under R version 4.0.5

## Loading required package: timeDate

## Warning: package 'timeDate' was built under R version 4.0.5

## Loading required package: timeSeries

## Warning: package 'timeSeries' was built under R version 4.0.5

library(tseries)

## Warning: package 'tseries' was built under R version 4.0.5

## Registered S3 method overwritten by 'quantmod':


## method from
## as.zoo.data.frame zoo
1. Get a feel of the data
str(MSFT_data) # structure of data

## 'data.frame': 121 obs. of 2 variables:


## $ Date : chr "11/1/2011" "12/1/2011" "1/1/2012" "2/1/2012" ...
## $ AdjClose: num 20.6 21.1 23.9 25.7 26.3 ...

class(MSFT_data)# check the class of the R object

## [1] "data.frame"

dim(MSFT_data) #check dimensions of the data

## [1] 121 2

head(MSFT_data)# view first few rows of the data

## Date AdjClose
## 1 11/1/2011 20.58750
## 2 12/1/2011 21.05067
## 3 1/1/2012 23.94554
## 4 2/1/2012 25.73761
## 5 3/1/2012 26.33147
## 6 4/1/2012 26.13558

tail(MSFT_data) # view the last few rows of the data

## Date AdjClose
## 116 6/1/2021 270.3824
## 117 7/1/2021 284.3656
## 118 8/1/2021 301.3032
## 119 9/1/2021 281.9200
## 120 10/1/2021 294.8500
## 121 10/8/2021 294.8500

2. Convert data into a time series object


MSFT_data_ts=ts(MSFT_data, start=c(2011,11),end=c(2021,8), frequency=12)

3. Plotting the price data


plot(MSFT_data_ts[,2], col="blue", ylab="Adjusted Closing prices", xlab="Time
",
main="Monthly closing prices of Microsoft")
legend(x = 'topleft', legend = 'MSFT', lty = 1, lwd = 2, col = 'blue')
hist(MSFT_data_ts) # plot histogram

CALCULATION OF RETURNS
1. Simple returns
n <- nrow(MSFT_data) #obtain the number of rows
close_prices <- MSFT_data$AdjClose #extract the closing prices
close_prices=as.data.frame(close_prices)
msft_ret <- ((close_prices[2:n, 1] - close_prices[1:(n-1), 1])/close_prices[1
:(n-1), 1])
# Add dates and plot
msft_ret1=ts(msft_ret, start=c(2011,12),end=c(2021,8), frequency=12)
plot(msft_ret1, col="red", ylab="simple returns", xlab="Time",
main="Monthly simple returns of Microsoft")

Alternatively use:
simplereturns<-returns(MSFT_data_ts, method=“simple”)
ts.plot(simplereturns)
2. Continuously compounded returns/log returns/geometric returns
msft_cret <- log(close_prices[2:n, 1]) - log(close_prices[1:(n-1), 1])
msft_ret2<-ts(msft_cret, start=c(2011,12),end=c(2021,8), frequency=12)
plot(msft_ret2, ylab="Log returns", xlab="Time",
main="Monthly log returns of Microsoft")
Compare simple and continuous returns
head(cbind(msft_ret, msft_cret))

## msft_ret msft_cret
## [1,] 0.022497486 0.022248151
## [2,] 0.137519110 0.128849672
## [3,] 0.074839502 0.072171350
## [4,] 0.023073861 0.022811685
## [5,] -0.007439462 -0.007467273
## [6,] -0.088382153 -0.092534404

#plot on the same graph


plot(msft_ret1, col = "blue", ylab = "Return",
main = "Monthly Returns on MSFT")
# Add horizontal line at zero
abline(h = 0)
# Add the continuously compounded returns
lines(msft_ret2, col = "red") # you can add legend
Descriptive statistics of returns
Use basicStats in fBasics
basicStats(msft_ret) # simple returns

## msft_ret
## nobs 120.000000
## NAs 0.000000
## Minimum -0.130247
## Maximum 0.196262
## 1. Quartile -0.007389
## 3. Quartile 0.056124
## Mean 0.024012
## Median 0.023392
## Sum 2.881475
## SE Mean 0.005223
## LCL Mean 0.013671
## UCL Mean 0.034353
## Variance 0.003273
## Stdev 0.057210
## Skewness 0.157667
## Kurtosis 0.571776

basicStats(msft_cret) #compound returns

## msft_cret
## nobs 120.000000
## NAs 0.000000
## Minimum -0.139546
## Maximum 0.179201
## 1. Quartile -0.007416
## 3. Quartile 0.054605
## Mean 0.022182
## Median 0.023123
## Sum 2.661783
## SE Mean 0.005101
## LCL Mean 0.012081
## UCL Mean 0.032282
## Variance 0.003123
## Stdev 0.055881
## Skewness -0.052434
## Kurtosis 0.460632

Alternatively have commands for individual tests


mean(msft_cret)

## [1] 0.02218152

var(msft_cret)

## [1] 0.003122672

stdev(msft_cret)

## [1] 0.05588087

skewness(msft_cret)

## [1] -0.05243391
## attr(,"method")
## [1] "moment"

length(msft_cret)

## [1] 120

Some common tests on return series


Q. Write the null and alternative hypothesis for each case
t.test(msft_cret) #testing mean return=0

##
## One Sample t-test
##
## data: msft_cret
## t = 4.3483, df = 119, p-value = 2.911e-05
## alternative hypothesis: true mean is not equal to 0
## 95 percent confidence interval:
## 0.01208063 0.03228241
## sample estimates:
## mean of x
## 0.02218152

adf.test(msft_cret)#testing for stationarity

## Warning in adf.test(msft_cret): p-value smaller than printed p-value

##
## Augmented Dickey-Fuller Test
##
## data: msft_cret
## Dickey-Fuller = -7.1404, Lag order = 4, p-value = 0.01
## alternative hypothesis: stationary

normalTest(msft_ret,method="jb") # testing for normality assumption

##
## Title:
## Jarque - Bera Normalality Test
##
## Test Results:
## STATISTIC:
## X-squared: 2.5073
## P VALUE:
## Asymptotic p Value: 0.2855
##
## Description:
## Wed Sep 20 14:45:11 2023 by user: ADMIN

Box.test(msft_cret,lag=12, type="Ljung")

##
## Box-Ljung test
##
## data: msft_cret
## X-squared = 17.343, df = 12, p-value = 0.1372

# testing for serial autocorrelation


Box.test(msft_cret^2,lag=12, type="Ljung") # squared return series. Compare w
ith price series

##
## Box-Ljung test
##
## data: msft_cret^2
## X-squared = 12.949, df = 12, p-value = 0.3728

In practice, the choice of m may affect the performance of the Q(m) statistic. Several values
of m are often used. Simulation studies suggest that the choice of m ≈ ln(T) provides better
power performance. This general rule needs modification in analyzing seasonal time series
for which autocorrelations with lags at the multiples of the seasonality are more important.
For instance, lags 12 and 24 are important for monthly time series
You can also check for serial autocorrelation using the acf and pacf plots
par(mfrow=c(1,2))
acf(msft_cret)
acf(close_prices) # check how the spikes pass through the confidence bands

pacf(msft_cret)
pacf(close_prices)
acf(msft_cret^2)
pacf(msft_cret^2)
Comment on the plots

You might also like