abhishek_89
Champion Member
- Feb 9, 2017
- 3,038
- Category........
- FSW
- NOC Code......
- 2173
- App. Filed.......
- 26-05-2017
- AOR Received.
- 26-05-2017
- Med's Done....
- 22-05-2017
- Passport Req..
- 24-07-2017
- VISA ISSUED...
- 04-08-2017
- LANDED..........
- 16-03-2018
Yo
You are missing the most important variable - no. of applicants entering the pool at each CRS rangeIt is just a linear regression model and 458 is just an artifact of historic data. It is just linear regression. So it bound to be rough and incorrect. In this case the model seems under-biased. You probably want to correct for the bias by adding a constant factor to the output.
Here is the dataset I ran it on
table { }td { padding-top: 1px; padding-right: 1px; padding-left: 1px; color: black; font-size: 11pt; font-weight: 400; font-style: normal; text-decoration: none; font-family: Calibri, sans-serif; vertical-align: bottom; border: medium none; white-space: nowrap; }
Days ITAs Cut-off
7 2902 468
7 3334 459
14 3508 453
14 3664 447
14 3611 441
7 3884 434
23 3749 441
12 3753 431
7 3923 423
7 3665 415
15 3796 423
13 3687 415
5 3877 413
28 3409 449
14 3202 440
21 3264 441
7 2991 433
14 3035 434
14 2772 435
14 2871 433
14 2801 438
14 2757 436
7 2000 458
7 2750 439
21 2750 452
cut-off = 0.4816750707*#of days - 0.01407997821*# of ITAs + 478.0315437
Slightly less under-biased model. I ran a multiple linear regression since we are trying to correlate 3 variables in the equation. Still not super accurate. Hope this helps.