And the snow circled and fell ...

    The MyClime project ( http://myclime.info or http://mycli.me ) continues to develop according to the plan and today I will tell you what new cities appeared on the service, ask your advice about precipitation and tell you how to use this service all the same :). MyClime is a service for analyzing weather forecasts from various sources and checking their accuracy. For each day, a history of forecasts is kept and, when this day comes, a reliability rating is calculated. Forecasts are downloaded from source sites, as is actual weather.

    So, we have information in the system for all cities with a million-plus population in Russia:
    1. Ekaterinburg
    2. Kazan
    3. Moscow
    4. Nizhny Novgorod
    5. Novosibirsk
    6. Omsk
    7. Rostov-on-Don
    8. Samara
    9. St. Petersburg
    10. Ufa
    11. Chelyabinsk
    In addition, information is also available on Minsk, Kiev, Odessa and a number of world capitals and major cities such as London, Rome, Berlin, Munich, etc.

    The second important innovation is precipitation forecasts for all forecast sources. Now you can see who predicted how much rain and when. Actually with precipitation, my need to listen to advice is connected. Now precipitation is not used in any way when evaluating forecasts, but it seems to me that they must be taken into account. And here the following questions arise - what weighting factors should the temperature and precipitation have when calculating the rating? For example, if the temperature accuracy for the day turned out to be 70%, and the precipitation only 20%, then it seems to me not very logical to summarize them with an equal coefficient of 0.5 - well, what is the difference +2 degrees or -2 degrees if I fell in the rain. On the other hand, a large coefficient for precipitation cannot be done either because that the difference between “little rain” and “rainy times” or “cloudy” and “cloudy” is pretty subtle and the main criterion of reliability will be “there was rain - there was no rain”. That's why I wanted to ask for advice as you, as users, would be more convenient. Maybe, for example, several ratings separately for precipitation, separately for temperature, but then again what to do for example with the wind - another number? It would be interesting to hear your opinion.

    And finally, the third important point - during the work a lot of questions from users have accumulated, and the usability of the site is at the level of the old, Soviet ZIL refrigerator, so I would like to answer everything with a brief instruction manual :) (By the way, if anyone knows an adequate designer who will be able to draw well both the site and the icon sets and is still set up for long-term cooperation, throw contacts please)

    So let's start - on the main page there is a list of available sources and the latest forecasts that they give. The city can be selected on the right in the drop-down list. If you want, then on the right you can change the display unit - Fahrenheit or Celsius.

    Opposite each source is its percentage rating. In order to calculate the rating of each source, I use an indicator such as reliability. In this case, the reliability of the forecast for any date X for N days, this is how much time for N days the forecast corresponded to the actual weather on that day. Those. if the forecasts for today at “today minus 5 days”, “today minus 4 days”, “today minus 3 days”, “today minus 2 days”, “today minus 1 day” were +1, +2, 0, -1 , +3, with actual weather 0, the reliability of the forecast from the source for today is 20%. (In reality, the formula is much more complicated, but if to a first approximation, then everything looks exactly that way).

    And the rating in this case is the average confidence over 5 days. The rating value can be used to understand whose forecast has been the most reliable in recent times, and whose value can be guided in the next few days.

    If you click on the link "Statistics for the month", then a page with information on the reliability for the last month will open. The pie chart shows the number of the most reliable forecasts for the last month for each source for a particular city. The histogram shows when and by whom the most reliable forecasts were made. This information can be used to make a decision on the use of the source in the long term (show your forecast on the site or use as a source) - several months or more.



    If you return to the main page and click on the temperature or precipitation value, a graph with the history of the forecast change will open. The right vertical axis corresponds to the date for which the story was made, and the graph shows how the value of the predicted temperature changed with approaching the day on which the forecast was made. For example, in the picture, the forecast made on March 3rd to March 10th showed -12 degrees, the forecast on March 5th for March 10th showed -7, etc. The blue area shows the actual temperature on March 10th. Thus, visually the accuracy of the forecast is the number of times the orange chart was inside the blue area. Well, just the answers to the most frequently asked questions: 1. Please add from the list of sources rp5.ru





    rp5.ru does not prepare its forecasts, but takes them from the British MetOffice service, the forecasts of which are used by the BBC. So rp5 forecasts are the same as BBC forecasts.

    2. Where do you get the actual weather values
    Actual values ​​are published by weather sites from various weather stations. I use this data to compile a picture of the actual weather.


    Also popular now: