Simple Sales Prediction
 
10_visualized
20160616

Visualized : plot

  • The black dots represent the monthly sales figures.
  • The blue line is the line fitted by R's lm() function through the sales dots.
  • The red dot is the predicted sale for month 13.
  • In each case the red dot lies on the blue line, which proves the correctness.

The R code

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25

    df=read.table("predicion.csv",sep=",",na.strings="null")

    df=df[!is.na(df$V3),]
    names(df)= c("shop","product","unit","month")


    dev.off()
    shops=unique(df$shop)
    products=unique(df$product)
    dev.new(width=8, height=16)
    par( mfcol=c(6,2) )
    for ( sh in shops) {
        for ( pr in products) { 
            y=df[ df$shop==sh & df$product==pr,"unit"]
            x=df[ df$shop==sh & df$product==pr,"month"]

            plot(x,y,pch=19,main=paste(sh,pr))
            fit=lm( y~x) 
            abline(fit,col="blue",lwd=3)

            xx=13
            yy=y[x==xx]
            points( xx,yy,pch=19, col="red",lwd=5)
    }}
 
Notes by Data Munging Ninja. Generated on nini:/home/willem/sync/20151223_datamungingninja/simplesalesprediction at 2016-06-25 10:02