Using Only Spark ML Library, Code the following 4 Steps

Using Only Spark ML Library, I just want to code the following 4 steps: i) Read in Monthly data from json format. ii) Convert Monthly format of data to Quarterly date format ii) Aggregate the Quarterly Data by Summing and Averaging them respectively, creating 2 series iii) Take Lag of the Quarterly data. That's all. Would you have time to show how to do this? Next, iv) Go back to step ii) just after it is done, and show how to: specifically remove Quarterly data before a certain date, say, 1Q 1995. Then take Log of the remaining Quarterly data, and repeat iii), ie Taking Lag on it. Send the 4 statements to me in a PySpark script. Prove it is working. I can provide you with monthly json data file if you need. (Note: So obviously, the whole purpose is just to read in Monthly json data and convert them to Quarterly format by summing and averaging respectively. And then taking Log on the Quarterly data remaining after a certain cutoff date, and creating a new series that is this one lagged by 1 period.) Term and condition: Milestone is Only released when project is delivered in Entirety to me, and proof of results being attained.