addressalign-toparrow-leftarrow-rightbackbellblockcalendarcameraccwcheckchevron-downchevron-leftchevron-rightchevron-small-downchevron-small-leftchevron-small-rightchevron-small-upchevron-upcircle-with-checkcircle-with-crosscircle-with-pluscrossdots-three-verticaleditemptyheartexporteye-with-lineeyefacebookfolderfullheartglobegmailgooglegroupshelp-with-circleimageimagesinstagramlinklocation-pinm-swarmSearchmailmessagesminusmoremuplabelShape 3 + Rectangle 1ShapeoutlookpersonJoin Group on CardStartprice-ribbonShapeShapeShapeShapeImported LayersImported LayersImported Layersshieldstartickettrashtriangle-downtriangle-uptwitteruserwarningyahoo

Traditions vs Statistics

  • Apr 26, 2016 · 7:00 PM
  • Restaurant Oase

Hi all,

I am glad to announce the next Zurich F# meetup, scheduled for Tue, 26 April at 7 PM, same place as usual. (If you come for the first time, please download this pdf to make sure you find the location.)

Natallie Baikevich, computer scientist and professional F# software engineer, has chosen a timely subject for her presentation:

Traditions vs Statistics

We'll soon find out what the next summer is gonna be like - Sechseläuten
is just around the corner. But how closely do the Böögg's forecasts
reflect the reality?
In this hands-on session we'll take a look at weather data and try to
understand it. Any languages and tools of your choice - all is fair.
Of course, there's a little starter pack for F# people, which will help to
dig into data, so no previous experience is required, but keep your
imagination ready and don't forget a laptop!

All the best and see you soon,

Marc

Join or login to comment.

  • Hermann

    (continuing the previous post, because there is a character limit on individual posts)

    Note that I only used the mean temperatures found in the data file "TG_SOUID100757.txt", using only rows where the quality (column "Q_TG") was indicated as valid (value "0"), converted the temperatures (column "TG") to whole degrees (dividing by 10) and using only dates that are in the summer (beginning 21st of June and ending 21nd of September, inclusive). I converted the Böögg duration to minutes (dividing by 60). Joining on the years returns a table containing both Böögg duration and average temperatures in the summer by year. I then plotted the table in Excel and calculated the correlation coefficient using the Excel function "CORREL", because I have no experience with R or other F# charting package yet.

    Hope this helps if any of you needed pointers to finish this exercise.


    Best regards,

    Hermann

    April 30

  • Hermann

    Hello everyone,

    I agree. That was a fun experience!

    I would like to thank Marc again for organizing the meetup and Natallie for preparing such an interesting hands-on lab exercise. I think the lab format works very well for F# meetups, because it allows you to actively get involved with F# and the corresponding topic. Also, working in groups is really great for teaching others that might not have that much experience in F#.

    I decided to finish the exercise today and got a correlation of 0.082759029. This matches the correlation of 0.08 mentioned in the official study that Natallie linked to :-). Best regards,

    Hermann

    1 · April 30

  • Marc S.

    Hi Natallie, thanks again for the fun meetup :)

    B.t.w., I finally found the way to convert to units of measure generically:

    [<Measure>] type PP
    let withMeasure<[<Measure>] 'T> n =
    LanguagePrimitives.Int32WithMeasure< 'T> n
    let pp = withMeasure<PP> 50

    Anytime again...

    1 · April 29

  • Natallie

    So, some statistics from your guesses:
    min: -1
    avg: [masked]
    median: -0.125
    max: 0.2
    actual: in your script ;)

    The link to my blog post from prev hear: http://type-nat.ch/traditions-vs-statistics-sechselaeuten/

    Official study (I found out about it only recently, still need to check it out...): http://www.meteoswiss.admin.ch/home/climate/past/climate-of-switzerland/reports-throughout-the-year/boeoegg-prognose.html

    And... http://tylervigen.com/spurious-correlations - some insightful correlations :)

    April 27

  • Shankar R.

    Hi Marc / Natallie, could I turn up without a laptop and hope to join someone in the group for XP? Mine has let me down since yesterday :) -Thanks, Shankar.

    April 26

    • Natallie

      Hi Shankar, I hope there'll be enough of them. The idea is to split into groups of 3+- anyway, so we should be fine.

      Natallie

      April 26

    • Marc S.

      Hi Shankar, you can use my Lenovo notebook. It has a U.S. keyboard ;)

      April 26

Our Sponsors

People in this
Meetup are also in:

Sign up

Meetup members, Log in

By clicking "Sign up" or "Sign up using Facebook", you confirm that you accept our Terms of Service & Privacy Policy