Go to page 1
  Forum » General » Further Economic Analysis... Date
Should I attempt to measure Inflation and Interest Rates?
Yes
No
608 msgs.
MVP of the game
 Ok Guys, I have ANOTHER idea

My g/f is away for 6 weeks (as of tomorrow), so I will have the time to at least halfway complete this project...

I want to CALCULATE and TRACK the rate of inflation in this game, alongside the interest rates.


So here's how it will work:

Interest Rates - I will note this down on a daily basis, and track it across the course of the season.

Inflation - Now this is a tough one. I will collate the prices of player sales on a weekly basis. That's the simple part. Here's where it gets a little complicated:

In a real economy, they calculate inflation using two different measures. The reason being, is that some measures calculate ALL goods (including fuel and utilities.) Other measures only calculate products you buy at retail. They also use these 'measures' because they use sampling in their data collection - Rather than track ALL purchases. We will attempt something similar here

Taking this example, we can interpret this in two ways. Either we separate our count into one that only measures auction prices, and one that includes hostiles and transfer agreements, OR, we separate our count to include a wider berth of players. (e.g. 75+ average.)

By measuring the weekly prices, and by working out the AVERAGE increases, we can then calculate the rate of inflation

Noting interest rates will help us understand the correlation (if any).

However, I may need help with this, and I DEFINITELY need advice on the sample range to use. e.g. LM 40/25 - What would be a suitable standard deviation from that? 38 - 42 average? 23 - 26 Age? Sounds simple, but what about a 38/26 LM or a 42/23 LM?

Simply put, we NEED a wide berth to get measurable figures. But too wide, and it simply won't work.


Thoughts?!?!?!


As usual, all feedback and help is welcomed!

Edited by Dee12345 26-03-2012 06:08

Edited by Dee12345 26-03-2012 06:09

Edited by Dee12345 26-03-2012 06:10 		26/03/2012 06:08
  - Div/Gr
Username
412 msgs.
First-team player
 Grabbing the data will be some what of a pain, probably easiest to use "similar transactions" function. Perhaps randomly pick 1 team from every division. Chart the similar transaction data for every player on their team over the course of time. This should give a wide variety of positions/ages/averages for the data set.

If you chart lowest price, highest price and average price that could also be interesting. This data would include both hostile and auction data. You could also pull it separately using the similar transaction data. 		26/03/2012 06:55
  - Div/Gr
Fiscal
2159 msgs.
Best scorer
 Since you're already wasting your time on here, why not? go for it

Edited by @Lifeguard 26-03-2012 07:39 		26/03/2012 07:39
  - Div/Gr
Username
3363 msgs.
Best scorer
 Use only auction prices - from a fiscal point of view, these are the only prices that mean anything.
Players that have more than one auction price (players sold a few times) in their history will be particularly useful.
You must include the top end players as well but just weight them in your calcs accordingly so they dont skew them too much. 		26/03/2012 07:47
  - Div/Gr
1382 msgs.
International
 Check out inflationdata.com for some info. 26/03/2012 14:41
  - Div/Gr
1382 msgs.
International
 So you would need to calculate two price indexs then? Lik, the very first time a player was sold at auction and the last time he was sold correct? Saying season 1 price and season #. Split the difference on age from 18-35 and go with the median which should be like a range from 25-27. But, this could be too old so get some more feedback, The more you narrow down the better your hypothesis of inflation will be.

Hostiles could cause a problem because you then have to account for relationship status between users i.e. A cf 48/23 is hostiled for 10mil and then auctioned for 17mil. .... Or .. A cf 48/23 is transfered by team A to team B for 12mil when an average price on auction is 19-21mil which gives you false data so auction is the best way as said before since it gives the end price.

While writing this on my ipad a thought occured that maybe you should track only Keepers since thats one position people always need and favored. It is also a position where the age becomes less of a factor and more about the averages.

Good luck , hope you get something for your efforts because this is going to take some hours and constant re-checking. 		26/03/2012 15:06
  - Div/Gr
Username
3880 msgs.
Best scorer
 Im losing what's left of my hair trying to sort out this world cup so as much as I would like to help in this huge project I feel as though I wouldn't be able to give it as much attention as I would like

Best of luck with it though buddy it's nice to see im joining a PL that has a huge interest in the game 		26/03/2012 20:44
  - Div/Gr
608 msgs.
MVP of the game
 Ok Guys, I have decided to take the following Positions:

GK
CDF
CF

Together, I shall group and average them according to the following criteria:

Average 40 - 42, Age 19 - 21

Average 40 - 42, Age 22 - 24


Average 44 - 46, Age 19 - 21

Average 44 - 46, Age 22 - 24


Average 50 - 53, Age 19 - 21

Average 50 - 53, Age 22 - 24


Average 60 - 65, Age 19 - 21

Average 60 - 65, Age 22 - 24


Every week, I shall take all the similar transactions involving these players. (Within the weekly timeframe)

I shall then note down the total cost of each category, divisible by the number of sales. This will be done for each position. (Average/Mean)

I shall then collate the information to provide overall inflation figures. This can also be utilised to demonstrate WHAT POSITIONS is driving up inflation, and WHEN.

I shall maintain both sets of statistics TWICE. Once using only auction prices, and Once using ALL transaction methods.

YES, I will appreciate any help in collating the data. Especially if anyone can help to design a Bot for all the data trawling required on a weekly basis - Even if the Bot only puts similar transfers into an excel spreadsheet!







Edited by Dee12345 28-03-2012 09:46 		28/03/2012 09:34
  - Div/Gr
Username
32 msgs.
Child's coach
 1. This should also be done with historical transfers too. If we have some idea of what inflation was like before and how it looks now then we could assess how well the anti-inflation measures are working.

2. Transfers seem to also vary based on other factors (time of season, day of week, time of day etc.) things like this should be logged as well as simply all the prices and then their effects can be removed from the analysis.

3. Good luck! 		31/03/2012 15:21
  - Div/Gr
     
Go to page 1
1