Posts filed under General (2829)

The International Statistical Institute is a professional organisation of statisticians. Its membership includes quite a lot of expertise on the production and use of official statistics, and it has a habit of commenting on political interference with official statistics.

The ISI has a statement on US official statistics

The International Statistical Institute (ISI) is concerned about government interference in the compilation and availability of Federal Statistics in the United States of America (USA). A proposed change in measuring and reporting U.S. Gross Domestic Product (GDP) serves as an important and timely example.

The statement goes on to explain that the USA has proposed removing government expenditures from GDP, and that this is both intrinsically a bad idea and also a weakening of international agreements on which official statistics to measure and how to define them.

[Disclaimer: I’m a member of the ISI, but I wasn’t involved in any way in producing this statement]

Team Ratings for Week 9

The basic method is described on my Department home page.
Here are the team ratings prior to this week’s games, along with the ratings at the start of the season.

Performance So Far

So far there have been 39 matches played, 29 of which were correctly predicted, a success rate of 74.4%.
Here are the predictions for last week’s games.

Predictions for Week 9

Here are the predictions for Week 9. The prediction is my estimated expected points difference with a positive margin being a win to the home team, and a negative margin a win to the away team.

Team Ratings for Round 6

The basic method is described on my Department home page.
Here are the team ratings prior to this week’s games, along with the ratings at the start of the season.

Performance So Far

So far there have been 40 matches played, 22 of which were correctly predicted, a success rate of 55%.
Here are the predictions for last week’s games.

Predictions for Round 6

Here are the predictions for Round 6. The prediction is my estimated expected points difference with a positive margin being a win to the home team, and a negative margin a win to the away team.

Team Ratings for Week 6

The basic method is described on my Department home page.
Here are the team ratings prior to this week’s games, along with the ratings at the start of the season.

Performance So Far

So far there have been 35 matches played, 25 of which were correctly predicted, a success rate of 71.4%.
Here are the predictions for last week’s games.

Predictions for Week 6

Here are the predictions for Week 6. The prediction is my estimated expected points difference with a positive margin being a win to the home team, and a negative margin a win to the away team.

• Coffee nerd James Hoffman has a good video on an issue that crops up a lot on StatsChat — totals vs means/rates — though in his case it’s about shorter vs longer espresso shots for making flat white/cappucino.
• From Axios: foreigners don’t seem to be visiting the US as much — this is more interesting than it sounds, since a lot of the dip must be short-notice cancellations, which are relatively difficult/expensive.
  
• According to the NZ Herald, Elon Musk wants zero tariffs between the US and Europe. The story doesn’t point out that was very close to the situation a week ago (eg, according to economist Justin Wolfers.)
• Friday’s rain pushed the Auckland dam water levels up nearly four percentage points (conveniently, the total storage is close to 100 GL, so 1 GL is about 1 percentage point), which was bit more than a third of the deficit from average.  While we’re being nitpicky, that should be 9.99% below, not “-9.99% below”
  
• New York has a new subway map
  

Tim Welch has produced a nice Auckland Cycleways Dashboard (it’s a draft at the moment, or ‘beta’ as we nerds say).

Here are some of the University-relevant routes (click to embiggen, as usual)

These are based on the Auckland Transport cycle counters, which are good for the city centre and major bike paths, but less informative for, say, people commuting in from the south, where there aren’t bike lanes or bike counters.

The tariffs announced by President Trump are based on what he describes as tariffs imposed by other countries on the US. These don’t bear any obvious relationship to the numbers that ordinary economists would describe as tariffs

However, a poster on Twitter, @nonagonono, found there is a strong relationship between the claimed tariff and the countries import/export balance with the US

We want to be a little sceptical of relationships that come just from data-dredging, with no real theory, but this relationship is tight enough that it probably isn’t just a coincidence, and probably does tell us something about the definition of ‘tariff’ the President is using.  If so, the ‘tariff’ of 20% that New Zealand allegedly imposes is actually a trade deficit of 20%, which is not quite what I get from NZ trade figures (24%) but is close enough that some slight changes in definition might explain it.

One News had a story based on a press release from NZ Food Network.  NZFN are a very worthwhile organisation. They coordinate donations and purchasing of bulk food for food banks, which they same amounts to about 45% of  the food being distributed.   If you were looking to donate to an organisation that helps feed people in this country, they look like a cost-effective choice. NZFN also run a food bank survey and a food security snapshot to find out about national needs and supply by food banks.

The headline figure in the press release is that the food banks “provided food to over 500,000 individuals each month”, based on the current survey. That’s a lot.

One News, but not the press release, said this was an all-time high.  Looking at previous annual reports and information sheets from NZFN, it doesn’t seem to be.  NZFN reported 454,000 in the first half of 2024, but their 2024 Annual Report said 655,152 for the second half of 2023 and the 2023 annual report says 480,104.  The number has been close to 10% of the New Zealand population ever since NZFN really got going. The scandal isn’t that the number is up, it’s that the number has been high for years!

What has changed noticeably is the total amount of food NZFN has been able to distribute.  The 2022 annual reports says about 8.6 million kg, the 2024 annual report says 5.6 million, and the 2024/2025 summer newsletter says 6.3 million.  They lost some government funding to purchase food, even as demand was increasing.  (If you look at the current press release you see a higher number that that for total food distributed, but I think that’s total distributed by the food hubs, rather than by NZFN)

I said “the number” is close to 10% of the population.  It’s a bit hard to find out exactly what this number means: is it genuinely unique people fed per month? Do the food hubs actually count how many people each recipient is feeding, and track individual visitors?  The reason I’m a bit suspicious about the number is a 2023 press release

Over half a million people are now being supported by NZFN food hubs each month but there is still a growing need. That works out to 1,868,491 more people being supported by NZFN, than they were in January 2020.

It’s not possible to get a total number of extra people from a monthly number, because you’d need to track individual people over the whole year; you can only get total number of visits.  Given this, it looks possible that that the 500,000 per month might also be total number of visits rather than unique monthly visitors.  It would still be a lot, either way.

You might think that I should answer this question by contacting NZFN rather than guessing.  I’m not sure I (as a One News reader) should have to, but I did actually try that.

Team Ratings for Week 15

The basic method is described on my Department home page.
Here are the team ratings prior to this week’s games, along with the ratings at the start of the season.

Performance So Far

So far there have been 112 matches played, 81 of which were correctly predicted, a success rate of 72.3%.
Here are the predictions for last week’s games.

Predictions for Week 15

Here are the predictions for Week 15. The prediction is my estimated expected points difference with a positive margin being a win to the home team, and a negative margin a win to the away team.

Team Ratings for Round 21

The basic method is described on my Department home page.
Here are the team ratings prior to this week’s games, along with the ratings at the start of the season.

Performance So Far

So far there have been 140 matches played, 105 of which were correctly predicted, a success rate of 75%.
Here are the predictions for last week’s games.

Predictions for Round 21

Here are the predictions for Round 21. The prediction is my estimated expected points difference with a positive margin being a win to the home team, and a negative margin a win to the away team.