Solved – Sales forecasting with non-stationary data

arimaforecastingstationaritytime series

I would like to do sales prediction based on my sales data for a particular product for a year. I understand this is non-stationary data which needs to be converted into stationary data and modeled using auto.arima() to do forecasting for the next 30 days. I would appreciate if anyone can help me on this using R as I am new to this domain.

Date       Code    Sales      
 2015-01-01 65323     2 
 2015-01-02 65323     1 
 2015-01-03 65323     3 
 2015-01-04 65323     3 
 2015-01-05 65323     2 
 2015-10-05 65323     1 
 2015-10-06 65323     1 
 2015-10-11 65323     1 
 2015-10-14 65323     1 
 2015-10-17 65323     2 
 2015-10-18 65323     1 
 2015-10-21 65323     1 
 2015-10-24 65323     1 
 2015-10-25 65323     2 
 2015-10-26 65323     1 
 2015-10-28 65323     1 
 2015-10-31 65323     2 
 2015-11-02 65323     1 
 2015-11-03 65323     1 
 2015-11-05 65323     1 
 2015-11-06 65323     1 
 2015-11-07 65323     1 
 2015-11-10 65323     2 
 2015-11-14 65323     3 
 2015-11-16 65323     1 
 2015-11-17 65323     1 
 2015-11-18 65323     1 
 2015-11-22 65323     3 
 2015-11-23 65323     3 
 2015-11-26 65323     3 
 2015-11-27 65323     2 
 2015-11-29 65323     4

Best Answer

Try to re-edit data into a doable way. You can manually type them in CSV file. Once you do it, use R studio to read the data and do the following (you can even copy and paste):

data <-  ts(XXXX,start=c(2005,01,01),frequency=365) # creating a time series object where XXX is the name of CSV file

growth <- diff(log(data)) #the way to handle non-stationary data. difference of log implements growth rate of sales

library("forecast", lib.loc="/Library/Frameworks/R.framework/Versions/3.2/Resources/library") # instilling forecast package. If you have not got it, download it.

model <- auto.arima (growth) #that is your model

forecast(model,h=30) #this will do a simple 30 days forecast

Note this is growth rate of sales. to get predicted sales figure, you need convert back.

Related Solutions

Solved – ARIMA forecasting with auto.arima() and xreg

Basically what caused the issue is the line ts(data[1:96,4],start=1, end=96,frequency =7 ), when you specify both start and end with frequency = 7, R is trying multiply the series so that it has a length of 96 weeks.

Recall R defines the start and end time in seasons (weeks in your case). Since you are fitting daily data, only specifying start = 0 or start = 1 should be sufficient.

Instead of running View(saledata), try to use saledata to debug yourself and you can see wrong length of time series is outputted .

Start = c(1, 1) 
End = c(96, 1)

When you do ARIMA forecast with xreg, basically you will need to create a matrix newxreg for your next 48 days with the same structure as xreg, then specify newxreg = newxreg in the forecast function. A good habit for the xreg and newxreg matrix would be to include a Day column that acts as an ordering for the data.

Time Series – Forecasting Technique for Daily Data with Monthly and Weekly Seasonality

You might want to look at http://www.autobox.com/pdfs/capable.pdf starting with slide 43 for an example and any number of my responses to this list as this subject has come up many times.

The issue is that DAILY DATA can be largely dependent on deterministic variables like day-of-the-week, week-of-the-year, month of-the-year, week-of-the-month, long-weekends, Fridays-before-a-Monday-Holiday or Mondays-after-a-Friday-Holiday and/or particular days-of-the-month effects.

A major hurdle for you is that holidays (before, on and after) are important and heuristics (i.e. not simply done!) are required to identify many of these structures. Furthermore there may be changes in daily patterns over time and different volatilities (uncertainties/variability) for different days of the week.

To determine these factors requires searching for patterns, not just fitting coefficients. Detecting level shifts and local time trends along with one-time unusual values is also critical beside correctly forming an appropriate structure (i.e. ARIMA model identification) is also critical as one needs to craft together a number of competing model possibilities. Finally changing error variance and/or changing model parameters over time need to be considered as they can come into play quite frequently.

In closing one needs to possibly bring user-specified predictor series and their lead and lag structure which may be needed to explain the series of interest.

Best Answer

Related Solutions

Solved – ARIMA forecasting with auto.arima() and xreg

Time Series – Forecasting Technique for Daily Data with Monthly and Weekly Seasonality

Related Question