Discussion of article "Exploring Seasonal Patterns of Financial Time Series with Boxplot" - page 26
You are missing trading opportunities:
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
Registration
Log in
You agree to website policy and terms of use
If you do not have an account, please register
...
In general, nothing is proved, it is only proved that if we take the average value of MA(25) and exactly on the 15-minute TF, then if we choose a criterion in the form of deviation of a few points, then without looking at the EURUSD chart, you can safely trade at night exclusively in SELL
Formulate WHAT is the proof of market regularity according to you. Otherwise, it is not clear why some people consider it a proof and others do not.
Formulate what is the proof of market regularity according to you. Otherwise, it is unclear why some people credit the proof and others do not.
open a topic, maybe I will join the discussion, but I will say at once that regularity is statistical data with formalised conditions of input data analysis, if we are talking about the material of the article - I don't want to discuss it anymore,
if we are talking about time ranges, the task should be clearly formulated, approximately as follows: here is BP, one of BP properties is binding of bars to the time of day, let's get statistics on time intervals of 1 hour, where are the subsequent bars (after the interval under study) in the future relative to the average value.....
and how the statistics was obtained with the help of GA tester, with the help of Boxplot, with the help of ZZ breaks is already the methodology, how much the methodology estimates this characteristic (where are the subsequent bars) is already shown by statistics. If statistical studies show that the given estimation works only at certain intervals, it means most likely correlation with other unexplored BP data, but not vice versa, that the study statistics has revealed a pattern here, and here it has not revealed a pattern there
open a topic, maybe I will join the discussion, but I will say at once that a pattern is statistical data with formalised conditions of input data analysis, if we are talking about the material of the article - I don't want to discuss it anymore,
if we are talking about time ranges, the task should be clearly formulated, approximately as follows: here is BP, one of BP properties is binding of bars to time of day, get statistics on time intervals of 1 hour, where are the subsequent bars (after the interval under study) in the future relative to the average value.....
and how the statistics was obtained with the help of GA tester, with the help of Boxplot, with the help of ZZ breaks is already the methodology, how much the methodology estimates this characteristic (where are the subsequent bars) is already shown by statistics. If statistical studies show that the given estimation works only at certain intervals, it means most likely correlation with other unexplored BP data, but not vice versa, that the study statistics has revealed a pattern here, and here it has not revealed a pattern there
I'm not talking about the material in the article at this point. I am interested in the concept of proof of market regularities. What can be considered proof and what can't. This is the starting point in arguing and evaluating the article.
I believe that:
"Proof" of market regularity is derived from the (unprovable) thesis of the cyclical nature of market dynamics and is built on "science-like" assumptions and probability theory. A certain pattern is chosen in advance, under which statistics is collected. Further, a pattern phenomenon is considered a regularity if it is repeated N number of times for N period and has links to other temporal phenomena. This is a conditionally acceptable "pseudo-proof".
The whole question is how nicely and competently it is presented in the article. In my opinion, this is a matter of preference.
I am not talking about the material of the article at this point. I am interested in the concept of proof of market regularities. What can be considered proof and what can't. This is the starting point in arguing and evaluating the article.
I believe that:
"Proof" of market regularity is based on the (unprovable) thesis of the cyclical nature of market dynamics and is built on "science-like" assumptions and probability theory. A certain pattern is chosen in advance, under which statistics is collected. Further, a pattern phenomenon is considered a regularity if it is repeated N number of times for N period and has links to other temporal phenomena. This is a conditionally acceptable "pseudo-proof".
The whole question is how nicely and competently it is presented in the article. In my opinion, this is a matter of preference.
hmmm, probably the first time I will say that you are right and have succinctly pointed out the problem of arguments, although my opinion has always differed from yours
hmmm, this is probably the first time I will say you are right and have succinctly pointed out the problem of arguments, although my opinion has always differed from yours
I have bolted on my OLAP to the bar analysis via the adapter for MqlRates and some other updates. For EURUSD M15 for the period from 2010 to 2019, I decided to calculate the ProfitFactor aggregator by Close-Open bar ranges, broken down by hours and days of the week. Since this aggregator gives the ratio of positive amounts to negative amounts, its maximum (greater than 1) and minimum (less than 1) values can be interpreted as suitable for buying and selling respectively (for selling, from the PF less than 1 shown, you need to take the inverse of 1/PF to get the profitability of selling). Here's the log (didn't make a graphic):
Each line has PF, hour and day of the week. Marked the most attractive options. You can see that it is recommended to sell at 23 and buy from 0 to 4 on almost all days.
1 - Hypothesis with some justification
2 - Statistical testing
3 - algorithm development
4 - Test in history and selection of parameters
5 - conclusions and perspectives
ALL
just this, almost 1-2-3 is in the article..... p4 was done by fxsaber ;-)
5 is missing
and between 4 and 5 there should be something else.
1 - Hypothesis with some justification
2 - Statistical testing
3 - algorithm development
4 - History test and parameter selection
5 - conclusions and perspectives
ALL
just that, almost 1-2-3 is in the article.... p4 was fulfilled by fxsaber ;-)
item 5 is missing
and between 4 and 5 there should be something else.