open All Channels
seplocked Market Discussions
blankseplocked T2 inflation a simple index
 
This thread is older than 90 days and has been locked due to inactivity.


 
Pages: 1 [2]

Author Topic

Cista2
Hydra Investment Fund
Posted - 2010.07.26 08:59:00 - [31]
 

July 26 data:

I indexed the 10-day and 1-month periods, i.e. compared the price today with the price 10 days ago and 1 month ago.

Within the 100 different items the following trends were seen:
Past 10 days: 53 % of items going down, 47 % going up
Past month: 34 % of items going down, 66 % going up


Comments: The 10-day index is now trending towards deflation, while the 1-month index has the strongest indication of inflation measured for any index period so far. It must be expected that the lack of price increase in the short term will spill over to the longer period next.

Link to the data collected.

Cista2
Hydra Investment Fund
Posted - 2010.07.29 07:29:00 - [32]
 

July 29 data:

I indexed the 10-day period, i.e. compared the price today with the price 10 days ago.

Within the 100 different items the following trends were seen:
Past 10 days: 57 % of items going down, 43 % going up

Comments: This is by far the lowest value that has been measured on a 10-day period. It now looks like the inflation period is going to cancel itself out.

Link to the data collected.

Cista2
Hydra Investment Fund
Posted - 2010.08.01 10:40:00 - [33]
 

August 1 data:

I indexed all periods, i.e. compared the price today with the price 10 days ago, 1 month ago etc.

Within the 100 different items the following trends were seen:
Past 10 days: 56 % of items going down, 44 % going up
Past month: 31 % of items going down, 69 % going up
Past 3 months: 46 % of items going down, 54 % going up
Past 6 months: 51 % of items going down, 49 % going up


Comments: The 1-3 months indexes showed a higher inflation tendency than last time they were measured. The 69% value is the highest value measured so far. The contrasting low 10-day value indicates that inflation will not intensify.

Link to the data collected.

Vasaczk
Posted - 2010.08.03 00:24:00 - [34]
 

Spaghetti chart with unlabled delta %'s imo Laughing

In fact I might see if I can do one, you just use the csv dump from metrics?

Cista2
Hydra Investment Fund
Posted - 2010.08.04 08:38:00 - [35]
 

August 4 data:

I indexed the 10-day period, i.e. compared the price today with the price 10 days ago.

Within the 100 different items the following trends were seen:
Past 10 days: 60 % of items going down, 40 % going up

Comments: Once again the lowest value that has been measured on a 10-day period, indicating the current inflation period will be cancelled out.

Link to the data collected.

Cista2
Hydra Investment Fund
Posted - 2010.08.07 07:28:00 - [36]
 

August 7 data:

I indexed the 10-day and 1-month periods, i.e. compared the price today with the price 10 days ago and 1 month ago.

Within the 100 different items the following trends were seen:
Past 10 days: 58 % of items going down, 42 % going up
Past month: 44 % of items going down, 56 % going up


Comments: The post-Tyrannis inflation period is coming to an abrupt end.

Link to the data collected.

Cista2
Hydra Investment Fund
Posted - 2010.08.09 10:11:00 - [37]
 

August 9 data:

I indexed the 10-day period, i.e. compared the price today with the price 10 days ago.

Within the 100 different items the following trends were seen:
Past 10 days: 65 % of items going down, 35 % going up

Comments: I couldn't wait 3 days to take this measure. And as I was half expecting, it looks like T2 prices are now dropping like.... some heavy free-falling object ugh

Link to the data collected.

Ford Chicago
Pandemic Legion
Posted - 2010.08.09 17:14:00 - [38]
 

Cista,

I find your methodology to be very weak because it is lacking scale and doesn't appear to take into account any market "noise". There is no scale on these numbers, so even if it is true that there is a downward trend, of what importance is it?

If the trend is .01 isk/week who cares?
If the trend is 1M isk/week it is clearly not sustainable, and thus it is an adjustment and not a trend.

What does your analysis look like if you allow for a 10% price buffer on each item?



Cista2
Hydra Investment Fund
Posted - 2010.08.09 17:46:00 - [39]
 

Originally by: Ford Chicago
There is no scale on these numbers, so even if it is true that there is a downward trend, of what importance is it?
It's not ideal. However, noone is providing any scaled data, which is why I came up with this index. It is very easy for me to monitor. Someone else is welcome to provide a better dataset, meanwhile this is the best we have.

My index is a measure of 100 items. I can make the assumption that there is a normal distribution of the *range* of change within the sample within a given week. If this assumption holds then the measure of *number* of items going up or down is also a measure of the scale of change per item.

In other words, if 75 % items are dropping, then the presumed normal distribution (bellshaped curve) of rate of change) will dictate that there is a good portion of the items that are dropping like crap, another portion that are more or less stable, and hardly any that will be increasing. Having all the items just decreasing by 1 % is extremely unlikely.

I haven't seen any hypothesis or data that would say that the rates are not normally distributed, and my measure should therefore be quite indicative.

Ford Chicago
Pandemic Legion
Posted - 2010.08.09 18:15:00 - [40]
 

Edited by: Ford Chicago on 09/08/2010 18:16:05
Edited by: Ford Chicago on 09/08/2010 18:15:41
Cista2: "I haven't seen any hypothesis or data that would say that the rates are not normally distributed, and my measure should therefore be quite indicative."

But what would make you suppose a Gaussian distribution in the first place? Your measure is over 100 items, which is certainly better than 10, but not as good as 1000. And the measure that you take is an absolute, as opposed to a percentile change. That is, it makes a much bigger difference if a 10 isk mod goes up by 1 isk than if a 1M isk mod goes up by 1 isk.

I find your work and analysis interesting, and I'm not just trying to be jerk, but I don't think that it provides a complete picture of Eve market dynamics. I doubt you intended it as such, so please don't feel that I'm denigrating your analysis or efforts.


Cista2
Hydra Investment Fund
Posted - 2010.08.09 18:23:00 - [41]
 

Originally by: Ford Chicago
please don't feel that I'm denigrating your analysis or efforts.
No problem, and if someone was publishing weekly data calculated from actual prices I would definitely follow that and let my index rest.

Cista2
Hydra Investment Fund
Posted - 2010.08.13 16:33:00 - [42]
 

August 13 data:

I indexed the 10-day and 1-month periods, i.e. compared the price today with the price 10 days ago and 1 month ago.

Within the 100 different items the following trends were seen:
Past 10 days: 72 % of items going down, 28 % going up
Past month: 51 % of items going down, 49 % going up


Comments: I would have sworn the 10-day index could not show such an extreme result. This could be the beginning of a very bad period for T2 - except for consumers Wink

Link to the data collected.

Cista2
Hydra Investment Fund
Posted - 2010.08.25 10:37:00 - [43]
 

August 25 data:

I indexed all periods, i.e. compared the price today with the price 10 days ago, 1 month ago etc.

Within the 100 different items the following trends were seen:
Past 10 days: 60 % of items going down, 40 % going up
Past month: 61 % of items going down, 39 % going up
Past 3 months: 45 % of items going down, 55 % going up
Past 6 months: 51 % of items going down, 49 % going up


Comments: Gloomy. The good news is in the larger perspective, that prices post Tyrannis have remained higher than before (as seen in the 3-month measure) - this might soon transcend into the 6-months measure as well.

Link to the data collected.

Cista2
Hydra Investment Fund
Posted - 2010.09.08 10:27:00 - [44]
 

Edited by: Cista2 on 08/09/2010 10:27:42

September 8 data:

I indexed the 10-day, 1-month and 3-month periods, i.e. compared the price today with the price 10 days ago, 1 month ago etc.

Within the 100 different items the following trends were seen:
Past 10 days: 56 % of items going down, 44 % going up
Past month: 57 % of items going down, 43 % going up
Past 3 months: 39 % of items going down, 61 % going up


Comments: The most remarkable result here is that over the 3 months that cover the post-Tyrannis period, most items have increased in price (61 being the highest value measured for a 3-month period)

Link to the data collected.

Cista2
Hydra Investment Fund
Posted - 2010.09.21 09:27:00 - [45]
 

September 21 data:

I indexed the 1-month, 3-month and 6-month periods, i.e. compared the price today with the price 1 month ago etc.

Within the 100 different items the following trends were seen:
Past month: 51 % of items going down, 49 % going up
Past 3 months: 40 % of items going down, 60 % going up
Past 6 months: 45 % of items going down, 55 % going up


Comments: Inflation has picked up again, and for the first time there is inflation trend on the 6-month data.

This was the last index with this exact set of items, next time I will be changing the items to correspond better with what I am trading in myself.

Link to the data collected.

Cista2
Hydra Investment Fund
Posted - 2010.09.30 06:52:00 - [46]
 

Edited by: Cista2 on 30/09/2010 06:55:54

September 30 data:

I indexed the 1-month and 3-month periods, i.e. compared the price today with the price 1 month ago etc.

Within the 100 different items the following trends were seen:
Past month: 56 % of items going down, 44 % going up
Past 3 months: 44 % of items going down, 56 % going up


Comments: This was the first measure with the "improved" index with new items, including ships also. In as far as they are comparable, both the 1-month and 3-month values took a dip compared to 8 days ago.

Link to the data collected.

Cista2
Hydra Investment Fund
Posted - 2010.10.08 09:24:00 - [47]
 

Still doing these from time to time. Today I made a measure over 6 months with the new index composition and it looks surprisingly bad, as seen in the link only 34 % of items have increased in price since before Tyrannis.

Cista2
Hydra Investment Fund
Posted - 2010.11.25 09:21:00 - [48]
 

I haven't done one of these in a while. Today I was surprised to see that T2 prices have held up well, in fact over the past 3 months there has been overall inflation.

Link to the data collected.


Pages: 1 [2]

This thread is older than 90 days and has been locked due to inactivity.


 


The new forums are live

Please adjust your bookmarks to https://forums.eveonline.com

These forums are archived and read-only