how to remove seasonality from a dataset

I have a dataset which consist of hourly wind speeds for fifteen years. When I used the autocorrelation on the wind speeds, it is clear that it contains a upward and downward trend as well as seasonality at every 24 lag. I have to make the dataset stationary by removing the trends and seasonality, I am new to matlab so I am not quite sure how to about doing this. Any help?

[vedesh Mohit]

could diff(data,24) work?

[Ashfaq Ahmed]

@vedesh Mohit Did you find an answer to this question? I have the same problem

Answer 1

[Image Analyst]

I'd break it out by day of the year so that you'd have 365 rows (one for each day of the year), and 15 columns (one column for each year). Then I'd average across columns to get the mean. Then I'd fit that to a model, such as a cosine. Then subtract the model from each data point to get the detrended value.