GrowthMPP {WALS}

Determinants of Economic Growth

Description

Growth regression data used in Magnus et al. (2010).

Usage

GrowthMPP

Format

A data frame with 72 observations on 11 variables:

country

factor. Name of the country.

gdpgrowth

Average growth rate of GDP per capita from 1960 - 1996 at purchasing power parity.

lgdp60

Logarithm of GDP per capita in 1960.

equipinv

Average real equipment investment share of GDP from 1960 - 1985 comprising investments in electrical and nonelectrical machinery (in relative prices constant across countries).

school60

Enrollment rate for primary education in 1960.

life60

Life expectancy at age 0 in 1960.

popgrowth

Average growth rate of population from 1960 - 1996.

law

Index for the overall maintenance of the rule of law ('law and order tradition').

tropics

Proportion of country's land area within geographical tropics.

avelf

Average of five different indices of ethnolinguistic fragmentation which is measured as the probability of two random people in a country not sharing the same language.

confucian

Fraction of Confucian population in 1970 and 1980.

Details

The dataset is used in Magnus et al. (2010) to illustrate the WALS model averaging approach and combines the data used in Sala-I-Martin et al. (2004) and Sala-I-Martin (1997). See the references for more detailed descriptions and original sources of the variables.

Source

WALS package for MATLAB (and Stata) provided on Jan Magnus' personal website. https://www.janmagnus.nl/items/WALS.pdf.

References

Magnus JR, Powell O, Prüfer P (2010). “A comparison of two model averaging techniques with an application to growth empirics.” Journal of Econometrics, 154(2), 139-153. doi:10.1016/j.jeconom.2009.07.004.

Sala-I-Martin X (1997). “I Just Ran Two Million Regressions.” The American Economic Review, 87(2), 178–183.

Sala-I-Martin X, Doppelhofer G, Miller RI (2004). “Determinants of Long-Term Growth: A Bayesian Averaging of Classical Estimates (BACE) Approach.” American Economic Review, 94(4), 813-835. doi:10.1257/0002828042002570.

Examples

## Replicate Table 2 in Magnus et al. (2010)
# NOTE: prescale = FALSE, still used old version of WALS in Magnus et al. (2010).
# Not recommended anymore!
fitMPP <- wals(gdpgrowth ~ lgdp60 + equipinv + school60 + life60 + popgrowth |
                  law + tropics + avelf + confucian, data = GrowthMPP,
                prior = laplace(), prescale = FALSE)
tableMPP <- cbind("coef" = coef(fitMPP), "se" = sqrt(diag(vcov(fitMPP))))
print(round(tableMPP, 4))