Technique to predict in which month an individual customer is more likely to transact

[Question]

Hi All,

I would like to predict in which month the individual customer is more likely to transact or to miss transactions. I have 2 years data (I have attached a sample). 

What is the right technique to use in this case? 

In case you can't see the attached spreadsheet, please see below how the data looks like..

Thank you so much for your help

customer_id	WEEK_START_DATE	WEEK_END_DATE	No_trans	Tot_sales
50026	30-Aug-15	05-Sep-15	.	.
50026	06-Sep-15	12-Sep-15	.	.
50026	13-Sep-15	19-Sep-15	.	.
50026	20-Sep-15	26-Sep-15	.	.
50026	27-Sep-15	03-Oct-15	.	.
50026	04-Oct-15	10-Oct-15	.	.
50026	11-Oct-15	17-Oct-15	.	.
50026	18-Oct-15	24-Oct-15	2	4
50026	01-Nov-15	07-Nov-15	1	2
50026	08-Nov-15	14-Nov-15	2	4
50026	15-Nov-15	21-Nov-15	3	6
50026	22-Nov-15	28-Nov-15	.	.
50026	29-Nov-15	05-Dec-15	.	.
50026	06-Dec-15	12-Dec-15	.	.
50026	13-Dec-15	19-Dec-15	1	2
50026	20-Dec-15	26-Dec-15	.	.

[Ksharp]

Poisson Regression Model ?

Check 

1) PROC GENMOD + REPEATED 

2)PROC GLIMMIX