Statistical estimation of EE enrty, what score @ when

[shasha1111]

I updated the modeling method. Now I simulate this problem with Monte Carlo. Link is still the same.
Comments on variables are welcomed!


[size=10pt]Current prediction is once the score is below 600, it takes about 10 months to go under 400. Good luck, guys~[/size]

Avg score # of draw min. score 1st draw here is in real the 5th or 6th draw.
380.7073 1.0000 454.6283
378.6770 2.0000 444.0503
376.8362 3.0000 436.6292
375.1845 4.0000 430.8353
373.5203 5.0000 426.2554
371.9667 6.0000 422.0814
370.4656 7.0000 418.4424
368.9880 8.0000 415.0694
367.5321 9.0000 412.0687
366.1085 10.0000 409.1746
364.6911 11.0000 406.5423
363.3290 12.0000 404.0040
361.9846 13.0000 401.6431
360.6232 14.0000 399.4153
359.3356 15.0000 397.2657

I am trying to post a google-doc l-----ink.....but not allowed to...
@ docs.google.com/spreadsheets/d/1o2LU_lsYEykl1UMOkiZF78ds0Q1XTJS1TdfUh8qaA8M/edit?usp=sharing @

The above is its google doc l---ink.
please leave comments so that I can perfect the model and make it close to real.
Data sources are that google spread sheet online[you all know that ].
I have not built a good enough model yet, considering in my model, the # of ppl in the pool will be 0 after a certain time period. But I can always adjust it, based on your suggestions!
Thanks
Have fun...

 

[atmtaatmta]

Re: Statistical estimation of EE enrty, what score @ when

The distribution of ppl in the pool is not normal.
It was close to sum of two normals before the first draw, now it is not.
While the distribution of newcomers can be considered to be sum of two normals, the distribution of ppl, who are left after each draw is ugly. Well, it still can be thought as sum of two normals without a head.

And how did you calculate number of ppl in the pool?

 

[shasha1111]

Re: Statistical estimation of EE enrty, what score @ when

The distribution of ppl in the pool is not normal.
It was close to sum of two normals before the first draw, now it is not.
While the distribution of newcomers can be considered to be sum of two normals, the distribution of ppl, who are left after each draw is ugly. Well, it still can be thought as sum of two normals without a head.

And how did you calculate number of ppl in the pool?

sum of two normal with same avg is still normal, though those parameters are different.
The avg I used is for those with max points of 600.

I assume that ppl with nomination and job offers will be cleaned up after next draw. I then add the # of people in the first 5 draws together for a total number of ppl with nomination and job offers(ppl w/t N&J). And then I applied the ration of ppl w/h N&J in that google spreed sheet to ppl w/o N&J, which is around 18.5. I round it to 20. This is how I get the # of people in the pool

 

[astro313]

Re: Statistical estimation of EE enrty, what score @ when

Hi Shasha ,
I have also done some estimations and the final results are a lot different. That's because there are a lot of variables here. Why do you think in the beginning (31-Jan) are around 70 kilo candidates in the pool? 8) As I read somewhere there were around 70 kilo applications but only 3 kilo were eligible, which means 3000 people were in the pool. Yes it seems to low, but that is what have been said until now.

representation of spreadsheet=ROS

As per my calculations , which are based in the EE spreadsheet which people have filled until now, considering different representations such as 20%,15%,10% and ~6% the results are different.
For example if representation of spreadsheet is 20% (excluding the +600 points candidates which for many reasons they didn't consider adding theirs selves in the sheet) after the 5th draw all candidates above 400 points will be selected if the number of candidates picked will be 1500 .
For the same conditions except the ROS , if ROS is 15% candidates above 400 points will be selected in the 6th draw.
For the same conditions except the ROS , if ROS is 10% candidates above 400 points will be selected in the 6th draw.
For the same conditions except the ROS , if ROS is 6% candidates above 400 points will be selected in the 8th draw.

 

[atmtaatmta]

Re: Statistical estimation of EE enrty, what score @ when

sum of two normal with same avg is still normal, though those parameters are different.
The avg I used is for those with max points of 600.

Alas, their averages are not equal. We have a peak at roughly 400 and at roughly 850.
I'm also not sure of the deviations as well. It isn't really the sum either, more like linear combination.
You can't just discard that, it renderes your model somewhat incorrect.
But that's true only for the first draw. Things become worse, when we look at other draws.

You always assume, that people scores a normally distributed. It is clearly incorrect. I'll try to explain in words.
For instance, you have a big pool of normally distributed candidates. You make a draw and take half of them. Thus, there is only half of a bell left. A couple of weeks pass, there are newcomers in the pool. They are distributed normally with the same average and deviation, but the are added only to a half of our bell. So, the distribution will look like half of a bell with little raise near it's highest point. It is in no way normal.

Another proof of mismodelling: from your table we can see, that the pool will become empty once ITA score reaches average. It is impossible, because half of the people have score below average (because avg=median in N). That means that once ITA score reaches 384, int the pool there will still be half of all candidates, who ever entered the pool.
Also, if in your model pool grows faster, than it is diminished by draws, your projected ITA will increase at each step. I am pretty sure, that the pool will always be growing before it stabilizes in couple of years due to common free market laws (people with low scores will be reluctant to enter it once it will be clear, that ITA points won't any longer drop significantly).

 

[atmtaatmta]

Re: Statistical estimation of EE enrty, what score @ when

Ah, you also estimate, that the pool grows by 2% per draw. How did you calculate this?
And why do think, that number of newcomers is anyhow connected to number of people currently in the pool?

 

[shasha1111]

Re: Statistical estimation of EE enrty, what score @ when

Alas, their averages are not equal. We have a peak at roughly 400 and at roughly 850.
I'm also not sure of the deviations as well. It isn't really the sum either, more like linear combination.
You can't just discard that, it renderes your model somewhat incorrect.
But that's true only for the first draw. Things become worse, when we look at other draws.

You always assume, that people scores a normally distributed. It is clearly incorrect. I'll try to explain in words.
For instance, you have a big pool of normally distributed candidates. You make a draw and take half of them. Thus, there is only half of a bell left. A couple of weeks pass, there are newcomers in the pool. They are distributed normally with the same average and deviation, but the are added only to a half of our bell. So, the distribution will look like half of a bell with little raise near it's highest point. It is in no way normal.

Another proof of mismodelling: from your table we can see, that the pool will become empty once ITA score reaches average. It is impossible, because half of the people have score below average (because avg=median in N). That means that once ITA score reaches 384, int the pool there will still be half of all candidates, who ever entered the pool.
Also, if in your model pool grows faster, than it is diminished by draws, your projected ITA will increase at each step. I am pretty sure, that the pool will always be growing before it stabilizes in couple of years due to common free market laws (people with low scores will be reluctant to enter it once it will be clear, that ITA points won't any longer drop significantly).

I see what you mean now. Actually, that is why I start real analysis after all top people(>600) are gone. So the rest would be only those with max pt of 600. Therefore, the distr. is close to one norm. Even if there are some >600 ppl later, their total number will be subtle, maybe 10-20ppl per draw? And I think you do not need to worry about such as small number at all.

I know the model was somehow wrong, haha, need to improve it.

You mentioned that the avg score of the pool will decrease.I see. I will try. That is a good idea.

 

[shasha1111]

Re: Statistical estimation of EE enrty, what score @ when

Ah, you also estimate, that the pool grows by 2% per draw. How did you calculate this?
And why do think, that number of newcomers is anyhow connected to number of people currently in the pool?

2% was a guess, no real support. More insightful numbers are welcomed!
A guess too. But I think in all, the avg of new incomers/draw should be a steady % of the pool each time.

 

[YamPower]

Re: Statistical estimation of EE enrty, what score @ when

Remember the PNPs that will be issued this year that gives 46,000 people 600 points.

 

[YamPower]

Re: Statistical estimation of EE enrty, what score @ when

By the way, if the EE numbers are sequential then there are over 92,000 people in the pool. My number is E000292xxx.

 

[fl_pie]

Re: Statistical estimation of EE enrty, what score @ when

By the way, if the EE numbers are sequential then there are over 92,000 people in the pool. My number is E000292xxx.

Thanks, good to know!

 

[Monija]

Re: Statistical estimation of EE enrty, what score @ when

2% was a guess, no real support. More insightful numbers are welcomed!
A guess too. But I think in all, the avg of new incomers/draw should be a steady % of the pool each time.

That is not true. The quota for PNP is 46 000, but just small amount of them will be eligible for 600 points in EE, other ones will get PR through regular paper application process. There will be around 3000-4000 PNP in EE.

 

[shasha1111]

Re: Statistical estimation of EE enrty, what score @ when

Remember the PNPs that will be issued this year that gives 46,000 people 600 points.

My understanding on it is the max # of ppl given pnp will be 46000. Next, not all of them are FSW, but can also be investment other catalog.

As mentioned in one of the ppl above, 'eligibal # of ppl in the pool is 3000', according to some article. I do not think there are only 3000 ppl eligible. But I do not think 90k would be a reliable guess, because some ppl might create multiple documents to apply and some opened the case, but failed to submit.

 

[shasha1111]

Re: Statistical estimation of EE enrty, what score @ when

Alas, their averages are not equal. We have a peak at roughly 400 and at roughly 850.
I'm also not sure of the deviations as well. It isn't really the sum either, more like linear combination.
You can't just discard that, it renderes your model somewhat incorrect.
But that's true only for the first draw. Things become worse, when we look at other draws.

You always assume, that people scores a normally distributed. It is clearly incorrect. I'll try to explain in words.
For instance, you have a big pool of normally distributed candidates. You make a draw and take half of them. Thus, there is only half of a bell left. A couple of weeks pass, there are newcomers in the pool. They are distributed normally with the same average and deviation, but the are added only to a half of our bell. So, the distribution will look like half of a bell with little raise near it's highest point. It is in no way normal.

Another proof of mismodelling: from your table we can see, that the pool will become empty once ITA score reaches average. It is impossible, because half of the people have score below average (because avg=median in N). That means that once ITA score reaches 384, int the pool there will still be half of all candidates, who ever entered the pool.
Also, if in your model pool grows faster, than it is diminished by draws, your projected ITA will increase at each step. I am pretty sure, that the pool will always be growing before it stabilizes in couple of years due to common free market laws (people with low scores will be reluctant to enter it once it will be clear, that ITA points won't any longer drop significantly).

as to the normal distribution, you are right. i am trying to mathematically figure out the potential distribution after taking away some top rated ppl. I guess, base on shape, chi would be a good choice. But i need to verify it first.

 

[YamPower]

Re: Statistical estimation of EE enrty, what score @ when

Interesting!

• Quebec does its own thing so EE won't select any provincial nominees for them.
  
• Manitoba will be issuing 5,000 PNPs but "Manitoba has been allocated a maximum of 500 additional nominations to directly select applicants for Manitoba from Express Entry."
  
• SINP (Saskatchewan) plans to take at least 9,550 in their program but only 1,000 through EE http://www.economy.gov.sk.ca/immigration/sinp-application-intake-thresholds
  
• Ontario NP plans to take 5,200 for 2015 but we don't know how many through EE because the program details haven't been announced
  
• Alberta NP can't be used for Express Entry
  
• PEI - Numbers couldn't be found
  
• Newf & Lab - NP Numbers haven't been announced but they took 579 in 2013
  
• New Brunswick - numbers couldn't be found
  
• BC has 5,500 places but how many will go through EE is unclear
  
• Nova Scotia - 350 through EE
  
• Yukon - Couldn't find figures on their NP
  
• NW Territories NP - 100 through EE