Testing real-time forecasting systems - page 42
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
Yeah, like you said: "the most important thing in our business is to wait." ^_^
I didn't say that, but the mechanic, one of the characters in the wonderful and beloved film "Old Men Fighting".
It's 5447 already. I'm proud of my posts and publications, terribly proud :)
P.S. Message to portal admins: please introduce an option to disable displayed rating for any user. I'll be the first to use this feature.
What are the shackles of glory heavy?
Found a bug in my new algorithm. I recalculated the forecast for Monday to Wednesday based on weekend data:
Calculated for 3 days.
Anyway, I have to check everything again, I mean the logic.
Sort of fixed it, the forecast for the weekend of 4 days was supposed to be like this:
Sort of fixed it, the forecast for the weekend of 4 days was supposed to be like this:
The forecast is timed to coincide - that's good enough.
The forecast is timed to coincide - that's good enough.
coincides in time - in the sense that I have accurately predicted a series of 15 minutes for 4 days ahead (1,2,3 ....)? :о)))
Objectively though - it has shown the main movement (nothing else needed from it, at least for now). I will not recalculate, let's see, theoretically it should grow from the next day. And it will be necessary to reconsider the calculation of RMS oscillations around the most likely trajectory. I notice that they are somehow underestimated, cautious. I was being "cautious", though, for simple safety reasons.
coincides in time - in the sense that I have accurately predicted a series of 15 minutes for 4 days ahead (1,2,3 ....)? :о)))
Objectively though - it has shown the main movement (I don't need anything else from it, at least not yet). I will not recalculate, let's see, theoretically it should grow from the next day. And it will be necessary to reconsider calculation of RMS of oscillations around the most probable trajectory. I notice that they are somehow underestimated, cautious. But I was being "cautious" for simple safety reasons.
Sergei. Without looking that far ahead (the whole trading week), what changed when you added the last two days (Monday, Tuesday), because prices went a figure lower from your calculations. The process itself is interesting. How far the trajectory will go with the new data.
What a load of rubbish you are writing, Helex. If you don't know what a terver is, don't say such nonsense.
I do not see anything delusional in my post, the usefulness of probability theory is questionable, in isolated cases it is useless. Its usefulness is only for casino owners and nowadays more DC.
I do not see anything delusional in my post, the usefulness of probability theory is questionable, in isolated cases it is useless at all. Its usefulness is only for casino owners and nowadays even DCs.
Prove it mathematically, if you're so smart.
xDDD
Very simple, just like in primary school. For example, we have a cellar full of apples, half of which are wormy and half not. It is impossible to determine the quality of an apple without biting into it (let's say). The apples have to be gathered in such a way that the number of good apples was greater than the number of real apples without having to bite into it. Every tenth good apple is eaten by Daughter Tsila (DC). But we are not stupid, we do the statistics so that next year we can gather more good apples than bad. And so on from year to year.
Attention is a question, in which year can we say with certainty that we will gather more good apples than bad? For the answer, see next year's textbook )))).
I'm not saying I'm so smart, I'm saying you're playing with casino odds no more. And it works for DC.