Bitcoin Forum

Economy => Economics => Topic started by: The Sceptical Chymist on December 27, 2018, 08:45:47 PM



Title: Merit source *ahem* needs help from math geniuses to evaluate a thread.
Post by: The Sceptical Chymist on December 27, 2018, 08:45:47 PM
Can someone please take a look at this thread (https://bitcointalk.org/index.php?topic=5089745) and the claims that the OP is making and then specifically take a look at this reply (https://bitcointalk.org/index.php?topic=5089745.msg48918433#msg48918433) of mine?  I've done a lot of chemistry mathematics in my life, but it's been a hell of a long time since I've done calculus and although I took a course in econometrics in college, that also was a long time ago and it didn't teach me much about TA and how calculus is applied to stock market analysis.  I suggested that the chart he presented wasn't inflation-adjusted and that the Dow components have changed over the years.  I'd like to hear some smart people's opinions on that.

Specifically, OP is claiming that the DJIA is linear from 1989-present and that
Adjusting the chart for inflation would only lower the slope of the linear trend. My point is not the % gain (the slope).
...and if I'm not mistaken, that's completely wrong, as I explained in my last post there.  I'd like to know if I'm correct from some people who know how to do serious math as it pertains to the stock market.  I know we've got a lot of brilliant folks here and I'd like to know if what I'm reading in the linked thread is a bunch of BS or if it has some validity.

Since Economics is full of spammers, this is going to be self-moderated.  Shitposts will get deleted, while thoughtful posts that teach me something will most likely earn some merit.  I'm a merit source but I'm not sure how many sMerits I have left.  If I run out and think someone who posted here deserves merits, I'll keep track and merit him when I get topped up.


Title: Re: Merit source *ahem* needs help from math geniuses to evaluate a thread.
Post by: mu_enrico on December 28, 2018, 03:31:05 AM
I'm not smart but have higher education in finance, so I also studied the stock market. The only fact stated by the OP is about the price will always go higher in the long run. It's not linear if you try to calculate it with regression analysis, the error (+-) rate will be too big.

The other claims have some fact behind it, but he twists his arguments to support his claims. Inflation-adjusted price is one method to isolate the inflation variable, but don't forget that there are too many variables affecting stock market. We had a difficult time to analyze causation versus correlation, as follows:

Wiseman says that the stock market condition is a reflection of a country economy. If a country (US) has a high GDP growth, it is just a simple consequence that the US stock market will rise. But what factors trigger the GDP growth? One of them is a low-interest rate so that printed money is used for productive activities. So what? a lower interest rate means a higher inflation rate.

It's not a "rigged market," just a consequence of government effort to increase the GDP growth.


Title: Re: Merit source *ahem* needs help from math geniuses to evaluate a thread.
Post by: STT on December 28, 2018, 03:25:35 PM
Well the basic argument is off because he is using face value of the dollar when it should be adjusted for inflation, some studies will do this.  I know my countrys currency halved in value over 20 years and now its even less, Im in europe supposedly we are rich but these figures over time will not be accurate or directly comparable.    FT will acknowledge such factors and maybe other serious finance sites :)

DXY is dollar index but its made up of similarly FIAT currencies, also devalued vs real things.    Another thing is government inflation is self adjusted by the concept of hedonics, which is the evaluation of worth in a product not just price itself rising.   The short take is the picture is not clear, market being rigged is nonsense so long as its open and apparent to all, price discovery is how markets set a price and doesnt mean accuracy is always available on that day.   WB calls this the MR. Market effect I think, prices like a supermarket can be bad or good and even over time can be wrong, overly speculative etc.  

Its an interesting argument either way.   Unfortunately there is too much politics influence in markets and not enough maths in there, but its not rigged or fake as everyone has that information

Quote
They print this money for bankers. That way you can't spend any of this money,

QE does hand money to primary dealers first.   The money will circulate but diminish in worth along with inflation.    Alot of QE is held by government pension funds, which do pay people the proceeds.   I kinda get his point, theres much inaccuracy in pricing

https://twitter.com/WallSt_Dropout/status/981270850048577537
https://i.imgur.com/C3UHTPS.jpg


Title: Re: Merit source *ahem* needs help from math geniuses to evaluate a thread.
Post by: The Sceptical Chymist on December 28, 2018, 08:24:08 PM
STT, I don't think you read the other thread in its entirety, much less all of my OP here that clearly stated that I'm looking for some mathematical analysis and a critique of where I'm right or wrong in that other thread and whether the OP of that thread doesn't know what he's talking about--and why, backed up with math. 

Most of what you wrote I already know.  I won't delete your post, because it's not a shitpost, but it's not what I'm looking for so I won't merit it.  But thanks for the input and bumping the thread for me.


Title: Re: Merit source *ahem* needs help from math geniuses to evaluate a thread.
Post by: STT on December 29, 2018, 06:19:00 AM
Thats ok if my posts are not relevant just delete, Im just posting with some personal interest but I asked a Maths major to take a quick look and his comments were
Quote
That linear line is complete trash for approximating that graph
the "percentage gain" doesn't have anything to do with the derivative, at least not like that
Just some 30 second feedback, I have seen this discussed previously but dont have a link for a proper article  on it to post right now.    The other point I would make is that income received from an asset must be included in its gains over time,  again FT.com used to show adjusted graphs for dividends etc but not every site will