Is any prediction doomed? - page 18

New comment
 
paukas:
Most pedestrian collisions statistically occur at pedestrian crossings.

The ratio of collisions to the total number of pedestrian crossings in and out of pedestrian crossings should be considered.
 
yeah... that's how understanding is tied up in terminology ;))))
 
paukas:
Most pedestrian collisions statistically occur at pedestrian crossings.

Yeah, and it's on a green light...
 
Debugger:

Good day to you all!
I've been doing forecasting algorithms for quite a long time, and I've achieved some results... Of course, there have been some ups and downs... Of course there were both ups and downs...
The more I am engaged in this business the more I have an idea that any forecasting is doomed

...

I would like to hear the opinion of forum users on this subject.

I am grateful in advance for comments, and for criticism too if there is any

. What is meant by "any prediction"?
 
paukas:
Most pedestrian collisions statistically occur at pedestrian crossings.
It would seem so! But I would slightly adjust "Most pedestrian collisions are observed to occur at pedestrian crossings." If "statistics" is synonymous with "observations".
 
yosuf:
It is not certain that it is doomed, but whether we know how to make predictions and whether we have laid the right assumptions in the basis of scientific research. It is necessary to search for regularities that increase the probability of the forecasts coming true. The search for regularities should not be opposed to forecasting or try to replace one concept with another.
Professor, you are the most competent person among us. You know how to arrange the minimum of words with the maximum of meaning very beautifully. Though I personally don't approve of your latest venture, could you please put your hand on your most precious possession and think about the answer to the question. It is relevant to the question at hand. Is it possible for a particular person to calculate the probability of successfully crossing a regulated pedestrian crossing?
 
HideYourRichess:
It would seem so! But I would slightly adjust "Most pedestrian collisions, according to observations, occur at pedestrian crossings." If "statistics" is synonymous with "observations".

By the way, the "observation results" can be used to predict, with certain tolerances, the number of collisions in the current reporting period

;)))) hell, what do pedestrians, traffic lights, turn signals and other nonsense have to do with it....

 
PapaYozh:

You have to look at the ratio of hit-and-runs to the total number of crossings in and out of pedestrian crossings.
And what would that yield? Isn't that the very probability that many here are eager to predict?
 
avtomat:

By the way, the "observation results" can be used to predict, with certain tolerances, the number of collisions in the current reporting period

;)))) what the hell do pedestrians, traffic lights, turn signals and other nonsense have to do with it....

Instead of words, try and you, calculate this probability, of your beloved successfully crossing the street on a pedestrian crossing. ;) then we can talk about forecasting. It should be much easier than counting probabilities in the market, shouldn't it?
 
HideYourRichess:
Instead of words, try to calculate the probability of successfully crossing the street on a pedestrian crossing. ;) then we can talk about forecasting.

I have better things to do.

It seems to me that you are simply unwilling to admit your mistake... I don't think I need to change your mind about anything.

Talking about forecasting, under some far-fetched and imposed conditions -- "first do the math, and then we'll talk" -- is bizarre, to say the least...

If you do not know the subject, I can suggest you some useful literature.

1...111213141516171819202122232425...43
New comment