Since 03/04/2004, JIT reporting became more regular than before and (I wish) more credible too. 
This gave me the opportunity to once again play a little with Excels capabilities of predicting dates based on the actual trend.
You can see my findings in the attached xls file.
Now, some explanations:
Sheet1
This sheet shows in the first two columns the JIT report date and the I485 date reported pairs since 03/04/2004.
The third and fourth columns show the same information PREDICTED by Excel BASED on the actual data from the first two rows. What it actually says is that, for example, if a JIT report is released on 7/2/2004, based on the latest trend, the I485 processing date will be at 4/8/2002.
Sheet2
Given the fact that Sheet1 looks very very optimistic, I decided to "adjust" a little the base data. I basically, predicted myself that the next two reports will come up with the same date of 1/30/2002 (as it happened with 1/9/2002). That way, I predicted myself the next two JIT reports and let Excel predict the rest based on the real JIT + my prediction. This time, as you can see, if a JIT report is released on 7/5/2004, based on the adjusted trend, the I485 processing date will be at 3/12/2002.
I also added the graphs that depict the actual/predicted trend for each case.
Of course, you can play further with this by adding more data in. For example, you can take in the consideration two more reports before 3/04/2004 which will make the prediction even less optimistic (and maybe more realistic). But I will let you folks do that.
To generate the prediction:
- make a copy of the base data (the actual trend)
- select all the data (both columns)
- RIGHT click on the bottom right corner of the selection
- WHILE keeping the right click pressed, drag the mouse down
- you will see your selection expanding over the empty cells below your data
- after selecting whatever # of empty cells (depend how far you wanna go with the prediction, release the right mouse button
- you will get a menu, select last option (Series)
- check the Trend checkbox and click OK
- you should see the data transformed in the predicted trend
NOTE: This stuff does not take in consideration the fact that the number of application is not the same for different periods of time, and that processing varies according to external factors (lazyness, screwness, joblessness, etc).
Have fun!
GG
This gave me the opportunity to once again play a little with Excels capabilities of predicting dates based on the actual trend.
You can see my findings in the attached xls file.
Now, some explanations:
Sheet1
This sheet shows in the first two columns the JIT report date and the I485 date reported pairs since 03/04/2004.
The third and fourth columns show the same information PREDICTED by Excel BASED on the actual data from the first two rows. What it actually says is that, for example, if a JIT report is released on 7/2/2004, based on the latest trend, the I485 processing date will be at 4/8/2002.
Sheet2
Given the fact that Sheet1 looks very very optimistic, I decided to "adjust" a little the base data. I basically, predicted myself that the next two reports will come up with the same date of 1/30/2002 (as it happened with 1/9/2002). That way, I predicted myself the next two JIT reports and let Excel predict the rest based on the real JIT + my prediction. This time, as you can see, if a JIT report is released on 7/5/2004, based on the adjusted trend, the I485 processing date will be at 3/12/2002.
I also added the graphs that depict the actual/predicted trend for each case.
Of course, you can play further with this by adding more data in. For example, you can take in the consideration two more reports before 3/04/2004 which will make the prediction even less optimistic (and maybe more realistic). But I will let you folks do that.
To generate the prediction:
- make a copy of the base data (the actual trend)
- select all the data (both columns)
- RIGHT click on the bottom right corner of the selection
- WHILE keeping the right click pressed, drag the mouse down
- you will see your selection expanding over the empty cells below your data
- after selecting whatever # of empty cells (depend how far you wanna go with the prediction, release the right mouse button
- you will get a menu, select last option (Series)
- check the Trend checkbox and click OK
- you should see the data transformed in the predicted trend
NOTE: This stuff does not take in consideration the fact that the number of application is not the same for different periods of time, and that processing varies according to external factors (lazyness, screwness, joblessness, etc).
Have fun!
GG