Data Forecasting (Wind Speed)

Dec 2015
3
0
Germany
Hi, I am engineer, therefore, know very little about the typical terminologies of stat and forecasting. Anyway, I tried to explain my prob. in simple word. I want to generate/forecast wind speed values. The data, I have are: I) Mean II) Max. and Min. differences from mean for the whole period III) Probability distribution of wind speed. Any suggestion or approach, please..
 
Jan 2015
20
2
Do you have a particular model in mind? Models can be like equations that describe the behavior of factors that you want to forecast.
 
Dec 2015
3
0
Germany
@mekun: Thanks for reply. Infact, I was thinking about first order auto-regression model (current value= last value*regression coefficient+noise function). However, the problem was of regression coefficient determination. I have one day mean hourly pattern, which I think can help to derive the coefficient. Anyway, this is my guess. If you can recommend alternative in the case, please...
 
Jan 2015
20
2
I usually use AR(1) models in econometrics, like forecasting economic growth, etc. I am not very familiar of its use in forecasting weather-related factors. Anyway, estimating the coefficient is easier with a statistical software. Eviews is the software I can readily think of. It has a very convenient time series tools. But I think, any statistical software that does regression can do it. It would be best to pick the software you are most familiar with.
 
Jan 2015
20
2
You may also want to check out agent-based modeling (ABM). I believe wind patterns operate in a very complex manner. Complexity is one of the areas being modeled in ABM.
 
Dec 2015
3
0
Germany
@mekum: Thank you very much...I try