Infographics Parades

When I saw Myko Clelland's tweet this morning, my reaction was "Wow! Just, wow!"

Myko (@DapperHistorian) kindly pointed me to the source of this photo that he tweeted about:

It appears on page 343 of Willard Cope Brinton's book, Graphic Methods for Presenting Facts (McGraw-Hill, 1914).

Myko included a brief description in his tweet, but let me elaborate by quoting from pp.342-343 of Brinton's book, and you'll see why I liked the photo so much:

"Educational material shown in parades gives an effective way for reaching vast numbers of people. Fig. 238 illustrates some of the floats used in presenting statistical information in the municipal parade by the employees of the City of New York, May 17, 1913. The progress made in recent years by practically every city department was shown by comparative models, charts, or large printed statements which could be read with ease fro either side of the street. Even though the day of the parade was rainy, great crowds lined the sidewalks. There can be no doubt that many of the thousands who saw the parade came away with the feeling that much is being accomplished to improve the conditions of municipal management. A great amount of work was necessary to prepare the exhibits, but the results gave great reward."

Don't you just love it? A gigantic mobile poster session!

© 2019, David E. Giles

A World Beyond p < 0.05

The American Statistical Association has just published a special supplementary issue of The American Statistician, titled Statistical Inference in the 21st. Century: A World Beyond p < 0.05.

This entire issue is open-access. In addition to an excellent editorial, Moving to a World Beyond "p < 0.05" (by Ronald Wasserstein, Allen Schirm, and Nicole Lazar) it comprises 43 articles with such titles as:

• The p-Value Requires Context, Not a Threshold (by Rebecca Betensky)
• The False Positive Risk: A Proposal Concerning What to do About p-Values (by David Colquhoun)
• What Have we (Not) Learnt From Millions of Scientific Papers With P Values? (by John Ioannidis)
• Three Recommendations for Improving the Use of p-Values (by Daniel Benjamin and James Berger)

I'm sure that you get the idea of what this supplementary issue is largely about.

But look back at its title - Statistical Inference in the 21st. Century: A World Beyond p < 0.05. It's not simply full of criticisms. There's a heap of excellent, positive, and constructive material in there.

Highly recommended reading!

© 2019, David E. Giles

The 2019 Econometric Game

The annual World Championship of Econometrics, The Econometric Game, is nearly upon us again!

Readers of this blog will be familiar with "The Game" from posts relating to this event in previous years. For example, see here for some 2018 coverage.

This year The Econometric Game will be held from 10 to 12 April. As usual, it is being organized by the study association for Actuarial Science, Econometrics & Operational Research (VSAE) of the University of Amsterdam. 

Teams of graduate students from around the globe will be competing for top prize on the basis of their analysis of econometrics case studies. The top three tams in 2018 were from  Universidad Carlos III Madrid,  Harvard University, and Aarhus University.

Check out this year's Game, and I'll post more on it next month.

(30 March, 2019 update - This year's theme has now been announced. It's "Climate Econometrics".)

© 2019, David E. Giles

Forecasting After an Inverse Hyperbolic Sine Transformation

There are all sorts of good reasons why we sometimes transform the dependent variable (y) in a regression model before we start estimating. One example would be where we want to be able to reasonably assume that the model's error term is normally distributed. (This may be helpful for subsequent finite-sample inference.)

If the model has non-random regressors, and the error term is additive, then a normal error term implies that the dependent variable is also normally distributed. But it may be quite plain to us (even from simple visual observation) that the sample of data for the y variable really can't have been drawn from a normally distributed population. In that case, a functional transformation of y may be in order.

So, suppose that we estimate a model of the form

              f(y_i) = β₁ + β₂ x_i2 + β₃ x_i3 + .... + β_k x_ik + ε_i ;    ε_i ~ iid N[0 , σ²] .                         (1)


where f(.) is usually a 1-1 function, so that f^-1(.) is uniquely defined. Examples include f(y) = log(y), (where, throughout this post, log(a) will mean the natural logarithm of 'a'.); and f(y) = √(y) (if we restrict ourselves to the positive square root).

Having estimated the model, we may then want to generate forecasts of y itself, not of f(y). This is where the inverse transformation, f^-1(y), comes into play.

Update for A New Canadian Macroeconomic Database

In a post last November I discussed "A New Canadian Macroeconomic Database".

The long-term, monthly, database in question was made available by Olivier Fortin-Gagnon, Maxime Leroux, Dalibor Stevanovic, &and Stéphane Suprenant. Their 2018 working paper, "A Large Canadian Database for Macroeconomic Analysis", provides details and some applications of the new data.

Dailbor wrote to me yesterday to say that the database has now been updated. This is great news! Regular updates are crucial for important data repositories such as this one.

The updated database can be accessed at www.stevanovic.uqam.ca/DS_LCMD.html .

© 2019, David E. Giles

Forecasting From a Regression with a Square Root Dependent Variable

Back in 2013 I wrote a post that was titled, "Forecasting From Log-Linear Regressions". The basis for that post was the well-known result that if you estimate a linear regression model with the (natural) logarithm of y as the dependent variable, but you're actually interested in forecasting y itself, you don't just report the exponentials of the original forecasts. You need to add an adjustment that takes account of the connection between a Normal random variable and a log-Normal random variable, and the relationship between their means.

Today, I received a query from a blog-reader who asked how the results in that post would change if the dependent variable was the square root of y, but we wanted to forecast the y itself. I'm not sure why this particular transformation was of interest, but let's take a look at the question.

In this case we can exploit the relationship between a (standard) Normal distribution and a Chi-Square distribution in order to answer the question.

Some Recommended Econometrics Reading for March

This month I am suggesting some overview/survey papers relating to a variety of important topics in econometrics:

• Bruns, S. B. & D. I. Stern, 2019. Lag length selection and p-hacking in Granger causality testing: prevalence and performance of meta-regression models. Empirical Economics, 56, 797-830.
• Casini, A. & P. Perron, 2018. Structural breaks in time series. Forthcoming in Oxford Research Encyclopedia in Economics and Finance. 
• Hendry, D. F. & K. Juselius, 1999. Explaining cointegration analysis: Pat I. Mimeo., Nuffield College, University of Oxford.
• Hendry, D. F. & K. Juselius, 2000. Explaining cointegration analysis: Part II. Mimeo., Nuffield College, University of Oxford.
• Horowitz, J., 2018. Bootstrap methods in econometrics. Cemmap Working Paper CWP53/18. 
• Marmer, V., 2017. Econometrics with weak instruments: Consequences, detection, and solutions. Mimeo., Vancouver School of Economics, University of British Columbia.

© 2019, David E. Giles