Supercomputer Ranking and Density

Top 500 Supercomputer Sites publishes a cool overview of the 500 most powerful (measured in Terra Flops) worldwide.

Here are the top 10 with five computers from the US, two from Germany and one each from China, Japan and Switzerland.

Looking at the whole list (500 super computers) the country distribution is as follows with the US dominating followed by China and Japan.

An a per capita basis, the US is still leading followed by Switzerland and Norway. The relatively low density of Germany and Japan may be surprising.

Patent Registration by Latin American Inventors and Applicants

The European Patent Office allows to search patent filing by a number of criteria, including country of inventor and applicant, publication data and classification (the search is for worldwide patent offices).

The following provides ans overview of Latin American (plus Jamaica, Trinidad & Tobago, French Guiana, Guayana and Suriname) countries in comparison with Spain and Portugal. Both the absolute number of filings and the filings per capita are shown (the data is shown for 2013 and the criteria was that at least one of the applicants and one of the inventors came from the respective country):

The leading country is Uruguay followed by Chile and Trinidad & Tobago. Compared to Portugal and especially Spain, there is still significant catching-up to do in Latin America (noting the Portugal and Spain also have a relatively low patent rate compared to the leading countries). The same information is shown graphically:

For Chile the development since 1990 has been as follows (significant increase since around 2000):

In terms of categories the distribution is as follows:

Luxury Car Density (2014 Update)

The following provides an update to an analysis we made in 2012, looking at the luxury car density across countries, posted here and here.

Again we looked at the number of listings on the OOYYO Network. This time we reduced the brands included only six main brands from Italy (Ferrari, Lamborghini and Maserati) and UK (Aston Martin, Bentley and Rolls Royce).

The density calculated as the listings on the site for the referenced brands divided by the total listings looks as follows (see map and table). The observations made in 2012 are repeated:

• very high densities for Luxembourg, Switzerland and Singapore
• high rates for UK and Italy
• medium-high rates for Western Europe, Sweden and Canada
• medium-low rates for Scandinavia (ex Sweden), US, Australia and New Zealand

Outliers include Lithuania, Malaysia, Slovakia and Nigeria. Note the some of these outliers might be driven by the fact that the density relative to all offered car is shown (and not per capita).

Without further comments, the map looks as follows:

And here the data table:

Preferred Podcasts

The following provides an overview of my favourite podcast (in English and German):

• Common Sense with Dan Carlin (US)
• Dan Carlin's Hardcore History (US)
• BBC Documentaries (UK)
• SRF Echo der Zeit (Switzerland)
• Economist: Editor's Pick (UK)
• EconTalk (US)
• SRF International (Switzerland)
• BBC More or Less: Behind the Stats (UK)
• Nature Podcast (UK)
• NPR Planet Money (US)
• NPR TED Radio Hour (US)
• SFR Trend (Switzerland)
• FT World Weekly (UK)

So far I have not encountered a podcast in Italian, Spanish or French which I found worth listening too (only referring to languages I would be able to understand).

Voter Turnout Chile - 2013 Update

We had covered in a previous post the development of voter turnout in Chile since 1870. In this post we update the time series to include the recent elections.

First we have update the population time series from INE and calculated the voting age population as 40% of the 15-19 age band plus all the age band 20+. The result is graphically depicted as follows:





















The ratio of the voting age population as a percentage of total population has increased from 64% in 1990 to 74% in 2013 and is expected to further increase at a somewhat reduced pace to 76% in 2020. It should be noted that the voting age population is not identical with the population permitted to vote as some segment of the voting age population (foreigner with less than five year of residency, persons with criminal records of certain characteristics etc) are not allowed to vote.

Separately, it is well known that the Servel voting registry is inflated and that the Census data contradicts the INE projection. We will come back to this in a future post.

Below updated table of (i) population, (ii) voting age population, (iii) votes and (iv) valid votes since 1870. The sources have all been disclosed in the earlier post.


































































And the voter turnout in graphical format.

Especially the first round of the presidential election seems to be in line with the previous compulsory voting record.

Impact of 1930´s depression on Chile´s Economy

Just re-read a chapter of Reinhart & Rogoff´s "This Time is Different: Eight Centuries of Financial Folly" (obviously I´m aware of the calculation errors, in my view it is still a tremendously valuable book).

The following graph (on page 267) describes the collapse of exports 1929 -1932. From a Chilean perspective the magnitude of the drop is particularly impressive.

Male/Female Ratio for Major Countries

For major countries, the total population and the male/female ratio has been obtained from UN´s: World Population Prospects: The 2012 Revision (dated 2013)

Major countries has been somewhat arbitrarily defined as:
- G20 member
- OECD member above 5 million population
- UN member abover 50 million

Graphically depicted this look as follows (red female surplus, blue make surplus):

The corresponding data in table format:

Copper Aluminium Price Ratio

We had looked at the copper to aluminium price ratio earlier and noted that it had reached a historical maximum. Interestingly the price ratio has more or less less stayed above 4 and recently even moved to 4.3 as can be seen in the following graph. The data has been obtained from OFDP (Open Financial Data Project).

Admittedly, this price development has not been in line with our expectations.

Proportional Voting Systems

There are a handful of different proportional voting and allocation systems, specifically

Quota Methods

The Hare Niemayer or Vinton is the largest remainder method, which is sometimes generalized with the use of different quotas:

The largest remainder method requires the numbers of votes for each party to be divided by a quota representing the number of votes required for a seat (i.e. usually the total number of votes cast divided by the number of seats, or some similar formula). The result for each party will usually consist of an integer part plus a fractional remainder. Each party is first allocated a number of seats equal to their integer. This will generally leave some seats unallocated: the parties are then ranked on the basis of the fractional remainders, and the parties with the largest remainders are each allocated one additional seat until all the seats have been allocated. This gives the method its name.

Hare quota (also called Hamilton): votes/seats 

Droop quota: 1+votes/(1+seats)

Hagenbach-Bischoff quota: votes/(1+seats)

Imperiali quota: votes/(2+seats)

The quota-method can also be used in an iterative approach, which mostly yields in the same result. An other alternative is too allocate all unallocated seats to the largest party.

Divisor Methods

The highest quotient and highest averages methods is defined as follows:

The highest averages method requires the number of votes for each party to be divided successively by a series of divisors. This produces a table of quotients, or averages, with a row for each divisor and a column for each party. The n'th seat is allocated to the party whose column contains the n'th largest entry in this table, up to the total number of seats available

The series of devisors are defined as follows (D'Hondt is also called Jefferson in the US and Hagenbach-Bischoff in Switzerland, Sainte-Lague is also called Webster in the US:

Generally D`Hondt (round-down) favours large parties while Adams (round-up) favours smaller parties while Sainte-Lague (artithmetic avergae), Hill Huntington (geometric mean) and Dean (harmonic mean) or generally almost identical in their outcome.

In a future post we will look at the actual implications.

Net International Investment Position (NIIP) 2011

The Net International Investment Position data for 2011 was retrieved from the IMF.

All countries with a (either positive or negative) balance larger than USD 100 billion are shown graphically:

Positive balance countries (data in USD billion). Top 5 are Japan, China, Germany, Switzerland and Hong Kong.

Negative balance countries (data in USD billion). Bottom 5 are United States, Spain, Australia, Brazil, Italy (surprising to see commodity exporters like Australia and Brazil with such negative balances).

Complete table (data in USD billion). The data is not consistent in itself as the total is about USD 1.5 trillion negative (a number of countries have no data but it is highly unlikely that they would add USD 1.5 trillion).



PS Chile has a balance of negative 24 billion.