Solved – Time series and multiple variables

exponential-familymultiple regressiontime series

I created a time series in Excel (not ideal) using Holts-Winters to forecast daily loan values in a month and it works very well.

I've been asked to build a similar model that integrates other variables but other than a multiple regression model (which has it's own issues) I'm not sure with my limited experience how I could achieve this and within Excel.

Any suggestions would be most welcome

Regards

Best Answer

The best way to do this is by using the Box-Jenkins Transfer function modeling approach as discussed in their text book called "Time Series Analysis".

You can create 6 dummies(ie 0/1 variable) representing the days of the week and 11 monthly indicators to represent seasonal effects. Now consider incorporating events such as holidays and include any lead, contemporaneous or lag effect around these known variables. Now there may be unusual values (pulses) or level shifts or local time trends in the data that you might need to identify and add these variables to the model so that the other variables have an accurate coefficient not skewed by outliers.

Furthermore the day-of-the-week effects may have changed over time e.g. there was no Saturday effect for the first 20 weeks but a Saturday effect for the last 50 weeks.If you wish to post your daily data we can try it.

Related Solutions

Solved – Time series analysis on login data to forecast CPU demand using R

Exponential smoothing is just a special case of an ARIMA model. If there is a benefit to fitting a general ARIMA model it is because of its generality and not that it handles gaps in the data any better than exponetial smoothing. I don't see any reason for throwing out February 29th. Individual dates would not have any appriciable effect on seasonality if there is some periodic component to the series. The time unit for time series analysis can be whatever time unit you measure the data in (it could be days weeks or years). You can cumulate data to create longer time intervals for the time series model. The fact that a date in one year falls on a different day of the week than in another year has nothing to do with its utility. If there are weekly effects this can show up in a 7 day periodic component. Gaps in the data does hurt your ability to fit the model. But if a single ARIMA model would have fit well to the complete series you probably can identify it piecing together the available portions of the series keeping count through the time index of the number of days missing at each gap. I don't understand why you can't have days as the time units. Is there a problem with having a long series because it seems to me that the time unit only affects the number of time points im the series?

Solved – Time series forecast in R with yearly frequency

Try using the tbats() function in the forecast package for R:

x <- ts(myts, frequency=365)
fit <- tbats(x)
plot(forecast(fit))

TBATS is a generalization of ETS models designed to deal with high frequency data. See http://robjhyndman.com/papers/complex-seasonality/ for the JASA paper behind it.

Best Answer

Related Solutions

Solved – Time series analysis on login data to forecast CPU demand using R

Solved – Time series forecast in R with yearly frequency

Related Question