Set Directory
getwd()## [1] "C:/Users/HP/Downloads/SDR_Datamatica/R_Directory"setwd("C:/Users/HP/Downloads/SDR_Datamatica/R_Directory")Install Required Packages
install.packages(c("xlsx","rJava","xlsxjars","XLConnectJars","XLConnect","lubridate","base"),repos = "http://cran.us.r-project.org",dependencies = TRUE)## Installing packages into 'C:/Users/HP/Documents/R/win-library/3.3'
## (as 'lib' is unspecified)## Warning: package 'base' is not available (for R version 3.3.1)## Warning: package 'base' is a base package, and should not be updated## package 'xlsx' successfully unpacked and MD5 sums checked
## package 'rJava' successfully unpacked and MD5 sums checked
## package 'xlsxjars' successfully unpacked and MD5 sums checked
## package 'XLConnectJars' successfully unpacked and MD5 sums checked
## package 'XLConnect' successfully unpacked and MD5 sums checked
## package 'lubridate' successfully unpacked and MD5 sums checked
##
## The downloaded binary packages are in
## C:\Users\HP\AppData\Local\Temp\RtmpUdKLSQ\downloaded_packagesinstall.packages("lubridate",repos = "http://cran.us.r-project.org",dependencies = TRUE)## Installing package into 'C:/Users/HP/Documents/R/win-library/3.3'
## (as 'lib' is unspecified)## package 'lubridate' successfully unpacked and MD5 sums checked
##
## The downloaded binary packages are in
## C:\Users\HP\AppData\Local\Temp\RtmpUdKLSQ\downloaded_packagesinstall.packages("base",repos = "http://cran.us.r-project.org",dependencies = TRUE)## Installing package into 'C:/Users/HP/Documents/R/win-library/3.3'
## (as 'lib' is unspecified)## Warning: package 'base' is not available (for R version 3.3.1)
## Warning: package 'base' is a base package, and should not be updatedinstall.packages("forecast",repos = "http://cran.us.r-project.org",dependencies = TRUE)## Installing package into 'C:/Users/HP/Documents/R/win-library/3.3'
## (as 'lib' is unspecified)## package 'forecast' successfully unpacked and MD5 sums checked
##
## The downloaded binary packages are in
## C:\Users\HP\AppData\Local\Temp\RtmpUdKLSQ\downloaded_packagesinstall.packages("zoo",repos = "http://cran.us.r-project.org",dependencies = TRUE)## Installing package into 'C:/Users/HP/Documents/R/win-library/3.3'
## (as 'lib' is unspecified)## package 'zoo' successfully unpacked and MD5 sums checked
##
## The downloaded binary packages are in
## C:\Users\HP\AppData\Local\Temp\RtmpUdKLSQ\downloaded_packageslibrary("xlsx")## Loading required package: rJava## Loading required package: xlsxjarslibrary("rJava")
library("xlsxjars")
library("XLConnectJars")
library("XLConnect")## XLConnect 0.2-12 by Mirai Solutions GmbH [aut],
## Martin Studer [cre],
## The Apache Software Foundation [ctb, cph] (Apache POI, Apache Commons
## Codec),
## Stephen Colebourne [ctb, cph] (Joda-Time Java library),
## Graph Builder [ctb, cph] (Curvesapi Java library)## http://www.mirai-solutions.com ,
## http://miraisolutions.wordpress.com##
## Attaching package: 'XLConnect'## The following objects are masked from 'package:xlsx':
##
## createFreezePane, createSheet, createSplitPane, getCellStyle,
## getSheets, loadWorkbook, removeSheet, saveWorkbook,
## setCellStyle, setColumnWidth, setRowHeightlibrary("base")
library("lubridate")##
## Attaching package: 'lubridate'## The following object is masked from 'package:base':
##
## datelibrary("forecast")## Loading required package: zoo##
## Attaching package: 'zoo'## The following objects are masked from 'package:base':
##
## as.Date, as.Date.numeric## Loading required package: timeDate## This is forecast 7.1library("zoo")read data from CSV
3.Importing the Excel file data into ‘R’
Customers_data_2013_2014<-read.xlsx("C:/Users/HP/Downloads/SDR_Datamatica/R_Directory/2013-16 Customer.xlsx",1)
Customers_data_2014_2015<-read.xlsx("C:/Users/HP/Downloads/SDR_Datamatica/R_Directory/2013-16 Customer.xlsx",2)
Customers_data_2015_2016<-read.xlsx("C:/Users/HP/Downloads/SDR_Datamatica/R_Directory/2013-16 Customer.xlsx",3)dim(Customers_data_2013_2014) ## [1] 76 9dim(Customers_data_2014_2015)## [1] 107 9dim(Customers_data_2015_2016) ## [1] 208 9customer_data<-rbind(Customers_data_2013_2014,Customers_data_2014_2015,Customers_data_2015_2016)
View(Customers_data_2013_2014)structure of the Date in cust dataset
str(customer_data$Transaction.Date)## Factor w/ 226 levels "01/01/2014","01/03/2014",..: 24 33 40 44 44 46 9 11 11 11 ...str(customer_data)## 'data.frame': 391 obs. of 9 variables:
## $ S.No. : Factor w/ 208 levels "1","10","11",..: 1 12 23 34 45 56 67 75 76 2 ...
## $ Value.Date : Factor w/ 239 levels "01/01/2014","01/03/2014",..: 24 34 42 46 46 49 9 11 11 11 ...
## $ Transaction.Date : Factor w/ 226 levels "01/01/2014","01/03/2014",..: 24 33 40 44 44 46 9 11 11 11 ...
## $ Cheque.Number : Factor w/ 1 level "-": 1 1 1 1 1 1 1 1 1 1 ...
## $ Transaction.Remarks : Factor w/ 378 levels "020021/CITI",..: 76 55 50 25 67 64 6 29 49 52 ...
## $ Withdrawal.Amount..INR..: num 0 0 111 10000 54213 ...
## $ Deposit.Amount..INR.. : num 20000 50000 0 0 0 ...
## $ Balance..INR.. : num 20000 70000 69889 59889 5676 ...
## $ NA. : logi NA NA NA NA NA NA ...names(customer_data)## [1] "S.No." "Value.Date"
## [3] "Transaction.Date" "Cheque.Number"
## [5] "Transaction.Remarks" "Withdrawal.Amount..INR.."
## [7] "Deposit.Amount..INR.." "Balance..INR.."
## [9] "NA."change the name of the Columns in cutomer data
names(customer_data)<-c("Si_No","Value_Date","Transaction_Date","Cheque_Number","Transaction_Remarks","Withdrawal_Amount_INR","Deposit_Amount_INR","Balance_INR")
View(customer_data)Ordering the Date format
customer_data<- customer_data[order(customer_data$Transaction_Date), ]——————-Perform Forecasting using ARIMA model———————
Extract Transaction_Date column from Dataset
customer_data<-ts(customer_data[,3],start = c(2013,1),end=c(2016,1),frequency = 12)
customer_data## Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
## 2013 1 2 3 3 3 4 5 6 6 7 8 9
## 2014 10 10 10 10 11 11 11 12 13 14 15 15
## 2015 16 17 18 19 20 20 21 22 23 23 24 25
## 2016 26Include the Plot Diagram
plot.ts(customer_data)
require(forecast)
ARIMAfit <- auto.arima(customer_data, approximation=FALSE,trace=FALSE)summary(ARIMAfit)## Series: customer_data
## ARIMA(0,1,0) with drift
##
## Coefficients:
## drift
## 0.6944
## s.e. 0.0768
##
## sigma^2 estimated as 0.2183: log likelihood=-23.18
## AIC=50.35 AICc=50.72 BIC=53.52
##
## Training set error measures:
## ME RMSE MAE MPE MAPE
## Training set 8.258254e-06 0.4543748 0.4129212 -0.4108704 5.127029
## MASE ACF1
## Training set 0.05135835 0.07142669pred <- forecast.Arima(ARIMAfit)
pred## Point Forecast Lo 80 Hi 80 Lo 95 Hi 95
## Feb 2016 26.69444 26.09573 27.29316 25.77880 27.61009
## Mar 2016 27.38889 26.54218 28.23559 26.09397 28.68381
## Apr 2016 28.08333 27.04634 29.12033 26.49738 29.66928
## May 2016 28.77778 27.58036 29.97520 26.94648 30.60908
## Jun 2016 29.47222 28.13346 30.81098 27.42477 31.51968
## Jul 2016 30.16667 28.70013 31.63320 27.92379 32.40954
## Aug 2016 30.86111 29.27707 32.44515 28.43853 33.28369
## Sep 2016 31.55556 29.86215 33.24897 28.96571 34.14540
## Oct 2016 32.25000 30.45387 34.04613 29.50305 34.99695
## Nov 2016 32.94444 31.05115 34.83773 30.04891 35.83998
## Dec 2016 33.63889 31.65319 35.62459 30.60202 36.67575
## Jan 2017 34.33333 32.25934 36.40733 31.16143 37.50524
## Feb 2017 35.02778 32.86909 37.18646 31.72636 38.32920
## Mar 2017 35.72222 33.48205 37.96239 32.29618 39.14827
## Apr 2017 36.41667 34.09787 38.73546 32.87037 39.96296
## May 2017 37.11111 34.71627 39.50595 33.44851 40.77371
## Jun 2017 37.80556 35.33701 40.27410 34.03024 41.58087
## Jul 2017 38.50000 35.95988 41.04012 34.61523 42.38477
## Aug 2017 39.19444 36.58472 41.80416 35.20322 43.18567
## Sep 2017 39.88889 37.21137 42.56641 35.79398 43.98380
## Oct 2017 40.58333 37.83970 43.32697 36.38730 44.77937
## Nov 2017 41.27778 38.46957 44.08598 36.98300 45.57255
## Dec 2017 41.97222 39.10091 44.84354 37.58092 46.36352
## Jan 2018 42.66667 39.73359 45.59974 38.18092 47.15241plot(pred,type="l",xlab = "Year",ylab = "Balance",lwd = 2,col = 'red',main="Forecasting using ARIMA Model")
Including Plots
Note that the echo = FALSE parameter was added to the code chunk to prevent printing of the R code that generated the plot.