python pandas OLS.predict, what is the correct signature?

Question

I am having a pandas OLS model,

mid_lag_lead_df_model

-------------------------Summary of Regression Analysis-------------------------

Formula: Y ~ <1> + <2> + <3> +

Number of Observations: 56 Number of Degrees of Freedom: 4

R-squared: 0.2864 Adj R-squared: 0.2452

Rmse: 0.0001

F-stat (3, 52): 6.9554, p-value: 0.0005

Degrees of Freedom: model 3, resid 52

-----------------------Summary of Estimated Coefficients------------------------

 Variable       Coef    Std Err     t-stat    p-value    CI 2.5%   CI 97.5%

---

        1     0.0527     0.0268       1.97     0.0547     0.0001     0.1053
        2    -0.0009     0.0269      -0.03     0.9730    -0.0537     0.0519
        3     0.0980     0.0268       3.66     0.0006     0.0455     0.1505
intercept     0.0000     0.0000       1.59     0.1184    -0.0000     0.0000

---------------------------------End of Summary---------------------------------

If i need to predict, the y value from [0.000207, -0.000361, -0.000091]

mid_lag_lead_df_model.predict

what the signature should be using predict as above?

Answer 1

You can do this to calculate y this your independent variables vector:

mid_lag_lead_df_model.beta.dot(pandas.Series([0.000207, -0.000361, -0.000091]))