OLS 回归与 statsmodels 给出许多系数

Question

I'm trying to make a simple regression on encryption time for a given datasize from a dataset. I'm a beginner with python and statsmodels but I think I'm getting strange results with OLS regression since it provides me a coefficient for every datasize like :

DataSize[T.1024] 0.0001
DataSize[T.1040] 0.0003
DataSize[T.1056] 0.0004
DataSize[T.1072] 0.0006
DataSize[T.1088] 0.0007

here is the code I developed:

    encrypt_key_16 = select_total_encrypt_time.loc[select_total_encrypt_time['KeySize'] == 16]
    y4, X4 = dmatrices('Measure ~ DataSize', data=encrypt_key_16, return_type='dataframe')
    mod4 = sm.OLS(y4, X4)
    result4 = mod4.fit()

Am I doing something wrong ?

Thank you in advance for your answer.

Answer 1

Okay I think I found what the problem is. When I print the dataframe X4 I'm getting the following output:

Output when printing X4

So I need to somehow make it only one column of DataSize.