Z-Score
Nov 22, 2010 at 11:57
6,535 Visualizações
13 Replies
Nov 22, 2010 at 11:57
(editado Nov 22, 2010 at 12:02)
I'm a bit confused about this. I didn't know what this means and had to look it up.
Is '0.26 (99.98%)' as it's supposed to be? The percents seem to indicate that the 0.26 should actually be calculated as 1/n = 3.85?
Or have I misunderstood this completely? Has this been 'turned around' so that the smaller the number the better the expectancy?
I found this chart:
Z-Score & Confidence Limit (%)
3.00 / 99.73%
2.58 / 99.00%
2.33 / 98.00%
2.17 / 97.00%
2.05 / 96.00%
2.00 / 95.45%
1.96 / 95.00%
1.64 / 90.00%
EDIT: Seems that everyone has got a 99.98%?
Is '0.26 (99.98%)' as it's supposed to be? The percents seem to indicate that the 0.26 should actually be calculated as 1/n = 3.85?
Or have I misunderstood this completely? Has this been 'turned around' so that the smaller the number the better the expectancy?
I found this chart:
Z-Score & Confidence Limit (%)
3.00 / 99.73%
2.58 / 99.00%
2.33 / 98.00%
2.17 / 97.00%
2.05 / 96.00%
2.00 / 95.45%
1.96 / 95.00%
1.64 / 90.00%
EDIT: Seems that everyone has got a 99.98%?
Less effort, better results.
forex_trader_7
Membro Desde Aug 01, 2009
941 posts
Nov 22, 2010 at 18:48
forex_trader_7
Membro Desde Aug 01, 2009
941 posts
Nov 22, 2010 at 19:03
I don't know. I was reading it but got distracted by a beer. Now I can't concentrate....
Membro Desde Jul 31, 2009
1449 posts
Nov 25, 2010 at 11:33
Guys, I've just noticed this topic.
Here's an explanation we've posted in our blog:
The sign doesn’t make the z-score value any better or worse – what we’re looking to see is a z-score value greater than 1.96 or less than -1.96 (Equal to 95% probability or better).
While z-score tells us if the streak of losses/wins are random or not, the sign simply tells us if the dependence is positive or negative, meaning:
Z-Score negative – dependence is positive.
Z-Score positive – dependence is negative.
Positive dependence – a profit will be followed by a profit and a loss by a loss.
Negative dependence – a loss will be followed by a profit and a profit by a loss.
And how certain can we be that it will happen? This is the percentage that is shown to you in your system’s z-score value. So for example, if z-score = 2.17(97.0%), it means there is a 97% confidence level that a loss will be followed by a profit and a profit by a loss (negative dependence). In such case you would want to miss a trade after a profit and take a trade after a loss to maximize profits and minimize losses.
Hope that’s clear enough! We’ll add a section describing it in full, in the next several days.
Here's an explanation we've posted in our blog:
The sign doesn’t make the z-score value any better or worse – what we’re looking to see is a z-score value greater than 1.96 or less than -1.96 (Equal to 95% probability or better).
While z-score tells us if the streak of losses/wins are random or not, the sign simply tells us if the dependence is positive or negative, meaning:
Z-Score negative – dependence is positive.
Z-Score positive – dependence is negative.
Positive dependence – a profit will be followed by a profit and a loss by a loss.
Negative dependence – a loss will be followed by a profit and a profit by a loss.
And how certain can we be that it will happen? This is the percentage that is shown to you in your system’s z-score value. So for example, if z-score = 2.17(97.0%), it means there is a 97% confidence level that a loss will be followed by a profit and a profit by a loss (negative dependence). In such case you would want to miss a trade after a profit and take a trade after a loss to maximize profits and minimize losses.
Hope that’s clear enough! We’ll add a section describing it in full, in the next several days.
Nov 25, 2010 at 11:41
Thanks for the explanation! I read the comments on your blog and the links too.
What is still confusing me is that the probability percent always seems to be 99.98%. On every system. There's a bug in the code? Or does it update at certain times only and gets calculated correctly later?
What is still confusing me is that the probability percent always seems to be 99.98%. On every system. There's a bug in the code? Or does it update at certain times only and gets calculated correctly later?
Less effort, better results.
Membro Desde Jul 31, 2009
1449 posts
Nov 25, 2010 at 11:51
Looks like an issue on our side - currently checking it, please stand by.
Membro Desde Jul 31, 2009
1449 posts
Nov 25, 2010 at 12:08
Ok, we've found an issue in our calculations - the fix is ready, however it will be applied only during the weekend as it involves a short period of downtime.
I apologize for any inconvenience.
Thanks for reporting it!
I apologize for any inconvenience.
Thanks for reporting it!
Membro Desde Jul 31, 2009
1449 posts
Jan 02, 2011 at 09:54
Siitari posted:
Z-Score (Probability): -3.53 (204.62%)
204 percent? :)
Shouldn't 100% mean that it will allways happen? I wonder how this is actually calculated...
Yes, statistically speaking, it can't be more than 100% 😄 - the fix is already ready so it will be applied during the next update.
You can refer here for a z-score calculator: https://faculty.vassar.edu/lowry/tabs.html
*Uso comercial e spam não serão tolerados, podendo resultar no encerramento da conta.
Dica: Postar uma imagem/URL do YouTube irá incorporá-la automaticamente no seu post!
Dica: Insira o sinal @ para preencher automaticamente um nome de utilizador que participe nesta discussão.
