Send output to:
Browser Blue - Charts White
Browser Black/White
CSV
Data:
95320.00 94965.00 94605.00 93860.00 101230.00 100840.00 95320.00 91650.00 92005.00 92005.00 92400.00 93110.00 94215.00 94215.00 93505.00 91650.00 101230.00 102690.00 100485.00 95320.00 97530.00 94215.00 95710.00 96425.00 97170.00 95320.00 95710.00 93110.00 101230.00 103795.00 101590.00 97530.00 101945.00 97170.00 101590.00 101230.00 102335.00 98275.00 102690.00 102335.00 108960.00 107465.00 101590.00 98630.00 102690.00 97170.00 101230.00 101945.00 103440.00 100130.00 101945.00 103050.00 107110.00 103795.00 99380.00 94605.00 99025.00 86875.00 92755.00 96065.00 99380.00 94605.00 94605.00 94605.00 97170.00 93505.00 88695.00 84670.00 87590.00 76190.00 83175.00 87235.00 87980.00 83920.00 84275.00 83175.00 86875.00 84275.00 79150.00 75445.00 81710.00 68105.00 76940.00 80965.00 80965.00 76190.00 71775.00 71420.00 75445.00 71775.00 64795.00 59985.00 65150.00 53005.00 64045.00 69920.00 71775.00 67715.00 62585.00 66255.00 67715.00 66610.00 55565.00 50440.00 54105.00 43065.00 54465.00 58525.00 61835.00 56315.00 51150.00 54105.00 55565.00 52645.00 41605.00 36795.00 41210.00 29065.00 42315.00 50440.00
Seasonal period
12
12
1
2
3
4
5
6
7
8
9
10
11
12
Type of Exponential Smoothing
(?)
Double
Single
Double
Triple
Type of seasonality
(?)
additive
additive
multiplicative
Number of Forecasts
12
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
Chart options
R Code
par3 <- 'additive' par2 <- 'Triple' par1 <- '12' par1 <- as.numeric(par1) if (par2 == 'Single') K <- 1 if (par2 == 'Double') K <- 2 if (par2 == 'Triple') K <- par1 nx <- length(x) nxmK <- nx - K x <- ts(x, frequency = par1) if (par2 == 'Single') fit <- HoltWinters(x, gamma=F, beta=F) if (par2 == 'Double') fit <- HoltWinters(x, gamma=F) if (par2 == 'Triple') fit <- HoltWinters(x, seasonal=par3) fit myresid <- x - fit$fitted[,'xhat'] bitmap(file='test1.png') op <- par(mfrow=c(2,1)) plot(fit,ylab='Observed (black) / Fitted (red)',main='Interpolation Fit of Exponential Smoothing') plot(myresid,ylab='Residuals',main='Interpolation Prediction Errors') par(op) dev.off() bitmap(file='test2.png') p <- predict(fit, par1, prediction.interval=TRUE) np <- length(p[,1]) plot(fit,p,ylab='Observed (black) / Fitted (red)',main='Extrapolation Fit of Exponential Smoothing') dev.off() bitmap(file='test3.png') op <- par(mfrow = c(2,2)) acf(as.numeric(myresid),lag.max = nx/2,main='Residual ACF') spectrum(myresid,main='Residals Periodogram') cpgram(myresid,main='Residal Cumulative Periodogram') qqnorm(myresid,main='Residual Normal QQ Plot') qqline(myresid) par(op) dev.off() load(file='createtable') a<-table.start() a<-table.row.start(a) a<-table.element(a,'Estimated Parameters of Exponential Smoothing',2,TRUE) a<-table.row.end(a) a<-table.row.start(a) a<-table.element(a,'Parameter',header=TRUE) a<-table.element(a,'Value',header=TRUE) a<-table.row.end(a) a<-table.row.start(a) a<-table.element(a,'alpha',header=TRUE) a<-table.element(a,fit$alpha) a<-table.row.end(a) a<-table.row.start(a) a<-table.element(a,'beta',header=TRUE) a<-table.element(a,fit$beta) a<-table.row.end(a) a<-table.row.start(a) a<-table.element(a,'gamma',header=TRUE) a<-table.element(a,fit$gamma) a<-table.row.end(a) a<-table.end(a) table.save(a,file='mytable.tab') a<-table.start() a<-table.row.start(a) a<-table.element(a,'Interpolation Forecasts of Exponential Smoothing',4,TRUE) a<-table.row.end(a) a<-table.row.start(a) a<-table.element(a,'t',header=TRUE) a<-table.element(a,'Observed',header=TRUE) a<-table.element(a,'Fitted',header=TRUE) a<-table.element(a,'Residuals',header=TRUE) a<-table.row.end(a) for (i in 1:nxmK) { a<-table.row.start(a) a<-table.element(a,i+K,header=TRUE) a<-table.element(a,x[i+K]) a<-table.element(a,fit$fitted[i,'xhat']) a<-table.element(a,myresid[i]) a<-table.row.end(a) } a<-table.end(a) table.save(a,file='mytable1.tab') a<-table.start() a<-table.row.start(a) a<-table.element(a,'Extrapolation Forecasts of Exponential Smoothing',4,TRUE) a<-table.row.end(a) a<-table.row.start(a) a<-table.element(a,'t',header=TRUE) a<-table.element(a,'Forecast',header=TRUE) a<-table.element(a,'95% Lower Bound',header=TRUE) a<-table.element(a,'95% Upper Bound',header=TRUE) a<-table.row.end(a) for (i in 1:np) { a<-table.row.start(a) a<-table.element(a,nx+i,header=TRUE) a<-table.element(a,p[i,'fit']) a<-table.element(a,p[i,'lwr']) a<-table.element(a,p[i,'upr']) a<-table.row.end(a) } a<-table.end(a) table.save(a,file='mytable2.tab')
Compute
Summary of computational transaction
Raw Input
view raw input (R code)
Raw Output
view raw output of R engine
Computing time
0 seconds
R Server
Big Analytics Cloud Computing Center
Click here to blog (archive) this computation