6,319 reputation
11139
bio website www54.sap.com/industries/…
location Switzerland
age 38
visits member for 4 years, 2 months
seen 13 hours ago

During the day, I forecast sales at supermarkets, drugstores, furniture, perfume and other retailers and calculate order proposals. Lots of time series, with an emphasis on fast, automatic and robust data cleansing and forecasting - with some logistical optimization thrown in for good measure. I'm active in the International Institute of Forecasters and an Associate Editor for their practitioner-oriented journal Foresight.

At night, I switch hats and do inferential statistics for clinical and biological psychology.

I'm never bored. And I use R.


1d
comment A better fit for sinusoidal data
If you don't have external causal variables that explain why a particular value at 10am is high rather than low, you won't be able to model it better. "Noise" is whatever you don't know about, and you apparently have a lot of it.
1d
revised Interpretation of mean absolute scaled error (MASE)
added discussion of forecast horizon
1d
awarded  Nice Answer
1d
comment Interpretation of mean absolute scaled error (MASE)
Related: stats.stackexchange.com/questions/124955/…
1d
revised Is it unusual for MEAN to outperform ARIMA?
added discussion about MASE
1d
revised Obtain the graph of the autocorrelation function in ARIMA models
added python tag
1d
comment Obtain the graph of the autocorrelation function in ARIMA models
This appears to be off-topic, as it is not about statistics as such, but "only" about the programming aspects. For this, StackOverflow.com is more suitable. However, there you will be expected to ask much more specific questions, "showing your work", and including a minimal reproducible example.
1d
comment Forecasting a solar data using arima in R
This appears to be off-topic, as it is not about statistics as such, but "only" about the programming aspects. For this, StackOverflow.com is more suitable. However, there you will be expected to ask much more specific questions, "showing your work", and including a minimal reproducible example. As it stands, this is a "do my job for me" question. I suggest you work through this extremely good resource.
1d
comment Is it unusual for MEAN to outperform ARIMA?
Thanks for accepting, but maybe you want to wait for a day - if a question has accepted answers, fewer people will even read it, let alone comment or answer. And other people may have different takes on this. Feel free to un-accept ;-)
1d
revised Arima time series forecast (auto.arima) with multiple exogeneous variables in R
typo
1d
comment Number of inputs used by ARIMA model
This may be helpful for how to do this in R: stats.stackexchange.com/questions/122803/…
1d
comment Number of inputs used by ARIMA model
I assume you are using the forecast package for R (as far as I know, there is neither a Forecast nor a forecasting package). Perhaps you could edit your question to clarify? (Sorry for nitpicking, but future readers will be grateful for precision.)
1d
answered Is it unusual for MEAN to outperform ARIMA?
1d
comment Range of Most Common Values
Does "not more than 10% different" mean that max<=1.1min, or min>=0.9max, or (max-min)<=0.1(max+min)/2?
2d
revised How to plot model with forecasts in R?
corrected code (airpass should be AirPassengers, an inbuilt dataset)
2d
comment Transform time-dependent data
@NickCox makes some extremely good points. On (B-)splines, I found Frank Harrell's Regression Modeling Strategies very helpful as an introduction. I don't have it here, but I seem to remember him discussing B-splines in particular, though not explicitly a periodic version (which you, as @NickCox notes, would need here), but this extension is completely straightforward. In R, look at bs() in the splines package.
2d
comment What is the distinction between short term and long term forecasting?
It certainly is not formally defined, at least not in the forecasting textbooks I am familiar with. Which is why I was so interested in examples for "short term forecasting methods". It appears to me like this distinction is used in a very ad hoc manner by different authors.
2d
awarded  Informed
2d
revised Recommendation for books/notes for linear mixed effect models for longitudinal data?
added references tag
2d
comment What is the distinction between short term and long term forecasting?
Careful: Zhanxiong talks about ARMA. Once you add differencing and move to ARIMA, you have trends, which are not dampened (as is frequently done in Exponential Smoothing). And of course AR(I)MA as such does not model external drivers, so I'd say Zhanxiong's answer looks at the problem from an orthogonal angle than I do.