Google


Saturday, March 31, 2007

FT on German growth

Bertrand Benoit writes in the FT about recent German economic success. Many French politicians take Germany as an example but forget the hard measures Germany had to go through, introduced by Schröder.

Benoit states that Germany seems to have overcome the backlog of unification by cutting labour costs. German growth is still driven by exports, although many German products are partly manufactured abroad. Germany seems to be the only contender to China in exports.

Thursday, March 29, 2007

Guido on Newsnight

Last night Guido Fawkes aka Paul Staines (or the other way around?) was on Newsnight and was heavily attacked by the Guardian's Michael White who gave him a good beating. Guido appeared without showing his face.

Guido admits that it didn't go right. Guido was mainly blamed for spreading conspiracies without proper foundation.

Nick Robinson defends himself against Guido's accusation that journalists are at politicians' mercy in his blog.

I very much agree with the journalists defence that Guido might stir things up but there is little substance behind his claims and rants.

Tuesday, March 27, 2007

nonlinear fitting

The New Economist links to a website which fits multidimensional functions to given data. As much as I agree that nonlinear function ought to be considered when analysing data, one should not forget that economics is a model-driven science, hence we should avoid data-mining.

Monday, March 26, 2007

real-time forecasting

The Fed in Philadelphia hosts a conference on Real-Time Data Analysis and Methods in Economics from April 19-20, 2007. One paper is by Philip Hans Franses, Erasmus University Rotterdam (with Dick van Dijk) titled Evaluating Real-Time Forecasts in Real Time. Dick van Dijk was my master thesis' supervisor.

Real-time forecasting becomes more and more evolved as automatic modelling and computation become more advanced.

Cameron gets the causality the wrong way.

Cameron gets the causality the wrong way.

But because marriage is a good institution that encourages people to commit to each other. The evidence is powerful. While half of unmarried couples split by their child's fifth birthday - the figure for married couples is one in twelve.

Obviously those who are unlikely to stay together will not marry in the first place.

Saturday, March 24, 2007

Award for Tong

The UK nonlinear news state

The Royal Statistical Society (UK) has announced in the March 2007 issue of the RSS News that the Guy Medal in Silver 2007 is awarded to Professor Howell Tong for his many important contributions to time series analysis over a distinguished career, and in particular for his fundamental and highly influential paper "Threshold autoregression, limit cycles and cyclical data", read to the Society in 1980, which paved the way for a major body of work in non-linear time series modelling. Previous recipients of the medal, which was inaugurated in 1893, include Maurice Bartlett, George Box, David Cox, Henry Daniels, David Kendall, C. R. Rao, Bernard Silverman, Adrian Smith, Peter Whittle, and others.

Screenwriters

According to IMDB Mario Puzo, Stephen King, Sergio Leone and JR Tolkien are the best screenwriters for the movies Godfather I and II; the Shawshank Redemption; The Good, Bad and Ugly and Lord of the Rings. I never understood why Shawshank Redemption rated so high, it's a good movie, but I fail to see the uniqueness, or marvellous acting?

Google high-fly


Google (GOOG) has had a return of 22% over the last year and has outperformed the NASDAQ (even) over the past six months. With a PE of 32 there is a question if the share is fairly priced. But with growth in Web 2.0 applications, Google stands firmer than Yahoo and Microsoft.

Bach


JS Bach is the greatest composer of the last millennium. Three days ago it was his 322th birthday and he is as alive as ever.

He has composed over 1000 pieces.

His well-tempered clavier is still the study book for many.

Many of the finest musicians devoted their lifetime to his music, eg Glenn Gould,

He was once called by the Zeit the fifth evangelist, as his music drags many people into church.

Cave on Letterman show

The great Nick Cave and his Bad Seeds ft. Chris Bailey, playing Bring It On at The David Letterman Show in June 2003

Senior Women

According Grant Thornton International and the Economist the Philippines has the highest proportion of women in senior management roles. Among the G7, the US has the highest rate with roughly 18%. Germany lags behind Britain and France. The Economist states:

the egalitarian legacy of communism could explain the high proportion of women near the top of companies in China and Russia.

Friday, March 23, 2007

Chinese shares

The Shanghai Stock Exchange issues the SSE 180 index, the five largest companies are (market capitalization in yuan):
  1. Industrial and Commercial Bank of China (1,397.86 billion)
  2. China Life Insurance (904.78 billion)
  3. Bank of China (887.31 billion)
  4. Sinopec (692.22 billion)
  5. China Merchants Bank (201.09 billion)
We see that the five largest listed companies are three banks, one insurer, and one oil company. ICBC is the second largest bank in the world after Citibank.

1 Chinese yuan = 0.129274126 U.S. dollars

Tuesday, March 20, 2007

stock market slump?





















Much has been said about the recent fall in stock markets. Compare the development of the FTSE100 with the two biggest shares in it - BP and HSBC - over the last 3 months.

Whereas the FTSE just regained its level of late 2006, BP fell considerably and HSBC fell somewhat.

Here is a list of the top 10 shares in the FTSE as of late 2006. Royal Dutch A and B together actually outweighs the other shares.

Economic news

According to Gerard Baker the chancellor's last budget will be less significant than what the Fed will announce tomorrow. Probably there won't be an interest rate change (fall), but many speculate about the hint of a recession, given recent financial stock market behaviour.

However, according to James Hamilton the probability of a recession in Q3 2006 was only 9%. As this is no forecast it still leaves the option of a recession open. According to Investment news:

However, economic forecasters in a survey released last week from New York-based WSJ.com put the odds of a recession over the next 12 months at just 25%.

Saturday, March 17, 2007

Spread Betting

Paton and Williams from Nottingham have investigated financial spread betting in ‘Quarbs’ and Efficiency in Spread Betting: can you beat the book?. They

examine a relatively novel form of gambling, index (or spread) betting, that mirrors (and indeed overlaps with) practices in conventional financial markets. In this form of betting, a number of bookmakers quote a bid-offer spread about the result of some future event, and bettors are invited to buy (sell) at the top (bottom) end of the quoted spreads.

This example explains betting in detail.

For example, in a cricket game between England and Australia, the bookmaker might set spread for runs in England’s first innings of 240-250. A bettor who believes England’s batting is weak may sell total runs at the price of 240 on a stake of, say, £5 per run. If England score 215 runs, this is the termination value of the asset and the bettor will win £125, calculated as the difference between the value and the price (240 - 215) times the stake (£5). On the other hand, if England score 290 runs in the game, the bettor would lose £250, calculated as the difference between 290 and 240 times the £5 stake.
They test two strategies:

We consider two alternative predictors of the actual value of the asset: (i) the average midpoint of the market bid-offer spreads (MID); (ii) the mid-point of the outlying bid-offer spread (OUTLIER).
And find:

Taken together, the evidence appears to be conclusive in suggesting that the market mid-point is systematically superior to the outlier in predicting actual assetvalues.
They also find

Using the notion of quasi-arbitrages or Quarbs, we find that it is possible to devise a trading strategy on the basis of the outlying spread that yields returns, both within and out of sample, that are consistently positive and superior to those that might be expected from noise trading.

Paul Fawkes

Guido Fawkes blog is one of the most popular political blogs in the UK. But why does the owner hide his name Paul Staines, who is a former trader? Is it to prevent others from defaming him, like he does with Labour politicians. This site reveals some of Staines' past.

Friday, March 16, 2007

Gary Koop

Gary Koop is a leading econometrician. He received his PhD from Toronto in 1989. Since then he has written many important papers, especially on Bayesian econometrics. He is currently professor at Strathclyde university - why he chose this nameless place I don't know.

Here is an example of his current work:
Forecasting in Dynamic Factor Models using Bayesian Model Averaging with Simon Potter, forthcoming in the Econometrics Journal

And one of his books:

Forecasting in Oxford

I went to a conference on forecasting in Oxford. Some researchers from the Lancaster Centre for Forecasting were presenting their work. Fildes and Goodwin were presenting evidence on sales forecasts by commercial companies.

It was startling to hear how forecasters in those companies were technophobe towards statistics and relied mostly on their own judgement, despite a poor track record. Goodwin explained some behavioural reasons for this fact. Apparently many sales forecasters would feel useless without meddling with the numbers. Also events were usually mis-anticiptated. Even confidence intervals would be seen as too advanced for them.

Some key lessons are:
• Don’t confuse your forecasts with your stock holding decision
􀀹 If you don’t know what your true forecast of demand is, you can never improve your performance.
• Forecasting accuracy is often not measured and monitored
􀀹 It should be!
• Benchmark your statistical forecast errors
􀀹 Some Statistical Systems forecasts are worse than useless!
• Understood why your forecasts are adjusted
􀀹 Too many small adjustments are made leading to no accuracy gains
􀀹 30% of adjustments are made in the wrong direction
• Consider developing your forecasting system to include
􀀹 Notes of why adjustments are made
􀀹 Summaries of the effects of similar adjustments, e.g. promotions

Thursday, March 15, 2007

US election odds

Here are the latest odds for the US election. H Clinton is the clear favourite, followed by Republicans Giuliani and McCain. 70 year old McCain looks like a too old option for me. Giuliani is not much younger at 62 but probably has more stamina.

Obama is fourth, and second Democrat. John Edwards and Al Gore are in fifth position, Al Gore not actually having declared his candidacy.

Tuesday, March 13, 2007

global warming swindle

I watched the great global warming swindle on C4 last night. Although biased in a totally different way than everyday news, it builds a strong case against the global warming hypocrisy. Its claims are:
  • the global warming movement is a platform for the left
  • global warming research attracts a lot of funding
  • the link between CO2 and global temperature is complex and causation is reverse
  • sunspots are more likely to cause warming
  • the IPCC is a governmental body censoring scientific opinion
  • the global warming agenda was initiated by Thatcher's desire to justify nuclear power
With which do you agree or disagree?

Monday, March 12, 2007

Randomness and risk

This article in HET discusses the concepts of risk and uncertainty in economics.

Much has been made of Frank H. Knight's (1921: p.20, Ch.7) famous distinction between "risk" and "uncertainty". In Knight's interpretation, "risk" refers to situations where the decision-maker can assign mathematical probabilities to the randomness which he is faced with. In contrast, Knight's "uncertainty" refers to situations when this randomness "cannot" be expressed in terms of specific mathematical probabilities.

Apart from this distinction - which can be disputed - there is also a difference between objective and subjective probabilities. Only with the work of von Neumann and Morgenstern was uncertainty formalised in the utility framework. From here it's only a small step to modern risk management in which once risks have been identified and assessed, all techniques to manage the risk fall into one or more of these four major categories: (Dorfman, 1997)
  • Tolerate (aka retention)
  • Treat (aka mitigation)
  • Terminate (aka elimination)
  • Transfer (aka buying insurance)

In financial risk management the measure of value-at-risk is often used. VaR is the maximum amount at risk to be lost from an investment (under 'normal' market conditions) over a given holding period, at a particular confidence level. This is shown in the graph below.

















Brown to enter on 5 July

Acoording to J Ashley in the Guardian Gordon Brown will take over as prime minister on 5th July.

The latest news I hear from the heart of the Labour party has Blair announcing his departure on May 10 or 11 and Brown kissing the Queen's hand (or whatever they actually do) on July 5.

Eurovision alliances or eurovisiopsephology


Soon it will be Eurovision again, this time from Helsinki. This paper by Gatherer in the Journal of Artificial Societies and Social Simulation analyses voting blocs in the contest. These are actually quite hard to ascertain but can be established by simulation. This graph of Venn diagrams, shows significant voting blocs over 6 or 4 year windows. Between 1996 and 2000 we have a Nordic bloc, and a Hungary-Malta-Slovenia-Macedonia bloc.

More recently we have
a huge Balkan bloc and a large Viking empire.

Saturday, March 10, 2007

Econometrician rankinng

On Peter Phillips' website we can find a (slightly outdated) ranking of economotricians by publications over the period 1989-1999. Phillips and Andrews from Yale are leading. Pesaran, Koop, and Steel are the only Europeans in the top 10.







1 Phillips, Peter C. B. Yale U 42
2 Andrews, Donald W. K. Yale U 29
3 Granger, Clive W. J. U CA, San Diego 29
4 Pesaran, M. Hashem Cambridge U 28
5 Diebold, Francis X. U PA 24
6 Rosenzweig, Mark R. U PA 23
7 Koop, Gary U Edinburgh 23
8 Steel, Mark F. J. U Edinburgh 23
9 Baltagi, Badi H. TX A&M U 23
10 Perron, Pierre Boston U 22

Obviously article is not article and so if we measure the impact we get a slightly different ranking. Andrews and Phillips swap place. The only European is Peter Robinson now.

Impact
rank

Author University

Number
of
articles

Impact
standardized
pages

All
econometrics
rank
/a
1989-1999

1

Andrews, Donald W. K. Yale U 29 526.262 2
2 Phillips, Peter C. B. Yale U 42 525.587 1
3 Heckman, James J. U Chicago 21 327.333 6
4 Robinson, Peter M. LSE 16 263.613 10
5 Perron, Pierre Boston U 22 262.085 4
6 Angrist, Joshua D. MIT 14 258.531 27
7 Rosenzweig, Mark R. U PA 23 253.765 13
8 Keane, Michael P. NYU 16 238.572 14
9 Horowitz, Joel L. Northwestern U 19 237.076 7
10 Stock, James H. Harvard U 22 231.003 11

LSE events

LSE hosts some interesting non-academic events, a list can be seen here. For instance on 28 March there is an event on Gordon Brown's future with Anthony Giddens (intellectual brain of New Labour), Tessa Jowell, Peter Riddell, Polly Toynbee and David Willets.

20 years Cointegration

End of march there is a conference in Rotterdam celebrating 20 years since the Engle and Granger paper introducing cointegration. Two time series are cointegrated if they are non-stationary but a linear combination of them is stationary.

Speakers include: Granger himself, Hendry, Pesaran, and Bruce Hansen. The program can be seen here.

Structural breaks

The most recent issue of Econometrica has an article by Qu and Perron called
Estimating and Testing Structural Changes in Multivariate Regressions. You can find a free WP version here. The study provides a unified treatment of models with common breaks, partial structural breaks models, block partial breaks models and locally ordered breaks models, among others. The latter is novel and of particular interest given potential wide applicability in contexts
related to the Lucas critique.

Rebels in the House

This blog (?) by some academics including Lord Norton publishes numbers on revolting backbenchers.

Baudrillard is merely a simulacrum


On Tuesday the philosopher Jean Baudrillard died at the age of 77. He was one of the leading media theorists. He described simulacra as a copy of a copy which has been so dissipated in its relation to the original that it can no longer be said to be a copy.

This article in the Welt lists some of his contributions and stresses Baudrillard's coolness.

Although heavily criticised the world is missing outstanding theorists of media entities.

Friday, March 09, 2007

London moving forward


According to PWC and the economist London will be the fourth biggest city economy after Tokyo, New York and LA in 2020. This is measured in GDP at PPP. London will surpass Chicago and Paris. Tokyo will lead with a hefty $1.6 trillion.

BBC on Germany

The relationship between Brits and Germans seems very much normal today, yet the reporting of Germany on British news is pretty awkward. For instance, this week a popular article was one about a couple of brother and sister who have children together (ie incest) living in East Germany. Another story reports about a trial involving a neo-Nazi burning Anne Frank's diary. There would be no problem if these stories would be as popular as stories about German politics, culture or sports. But they are not, the odd perversity from Germany seems more interesting than real life.

Wednesday, March 07, 2007

Rattle in Barbican



We heard Simon Rattle in the Barbican tonight. His Berlin Philharmonics were playing Dvorak, Ades, and Janacek. Tomas Ades is a young British composer whose piece Tevot had his UK debut. The piece was backed by a full orchestra, in Strawinksy style.

This article in the tagesspiegel reviews the German premiere. Probably a bit harsh on Ades, it hit the point that Dvorak was the strongest performance of the evening, and in a long time.

Tuesday, March 06, 2007

SAS programs

On this website you can find some SAS programs which I have written.

Customer Value Calculator

On this website you find a Customer Value Calculator. It was written by Calciu of Lille and Reinartz form INSEAD. It calculates the profit given retention rates and marketing costs. It is a valuable tool for Customer Relationship Management (CRM).

Monday, March 05, 2007

Wrong guy

Many might have seen this, here is a guy who comes to BBC for a job interview and they put him for a TV interview. I like his face in the beginning

Sunday, March 04, 2007

Trombonist


Christian Lindberg is the (technically) best and most eccentric trombonist living today. The Swede was born in 1958. He regularly appears in costume on stage and directs, as well as acts.

Many composers write special pieces just for him.

I saw/heard Lindberg in 2003 in Helsinki where he premièred his own work Helikon Wasp.

Friday, March 02, 2007

Job wanted

Want to employ an analyst in London, see this advert.

Jenkins on the Olympics


Simon Jenkins complains about the Olympics and their rising cost, currently estimated at 10 billion pounds.

Why are the British so critical about the major sports event? It would put London in the spotlight and create great infrastructure. Many people discuss the costs as if they all fell into a deep hole. The infrastructure remains there after the event.

How could small Greece afford to stage such events - well admittingly not being honest about the costs - and Britain doesn't dare?

I have to admit there is a big risk of wrong investment, shown by the Dome and Wembley, but pessimism won't get London anywhere.

Thursday, March 01, 2007

Polls prediction

Despite the recent polls showing that Cameron would pull many votes the electoral calculator predicts that the Tories would not gain a majority, nor would Labour, hence a hung parliament.

Party2005 Votes2005 SeatsPred VotesPred Seats
CON33.24%21136.92%288
LAB36.21%34632.45%291
LIB22.65%6419.54%36

Tax breaks for married couples

David Cameron tries to fix society by re-introducing tax allowances for married couples.

It is clear that it is not the role of the tax benefit system to make people marry. A tax incentive surely won't keep a broken family together.

What the debate also neglects is that the tax allowance creates a disincentive for the second earner (typically the wife) to work. It drives up the marginal tax rate for the second earner.

Interestingly, academics blame the reduction in male employment for the increase in single households. This would imply that Cameron should support the male breadwinner model, rather than tax allowances.

I agree that middle class families should keep more of their money but this can be achieved in more efficient ways.