Steve Bond, Asli Leblebicioglu, and Fabio Schiantarelli, "Capital Accumulation and Growth: A New Look at the Empirical Evidence", Journal of Applied Econometrics, Vol. 25, No. 7, 2010, pp. 1073-1099. The zipped folder "bls_datarep" contains five data sets that were used in the estimations. All files are tab-delimited text files. The zipped folder "bls_programrep" contains the Stata do-files that replicate the results in the tables. Note that the data sets need to be converted to Stata format before running the do-files, and path statements may need to be changed. All files are in DOS format; Unix/Linux users should use "unzip -a". The file descriptions are as follows: table1a.do: generates the ADF test results in Table 1A table1b.do: generates the Im-Pesaran-Shin unit root test results in Table 1B table1c.pgr: (RATS file) generates the Pedroni cointegration test results in Table 1C table2.do: generates the results of the pooled annual models in first-differences in Table 2 table3.do: generates the results for the robustness checks in Table 3 table4.do: generates the results of pooled annual models for the OECD and Non-OECD subsamples in Table 4 table5fs_all: generates the mean group estimates for the full sample in Table 5 table5fs_oecd: generates the mean group estimates for the OECD subsample (subtracting full sample means) in Table 5 table5fs_nonoecd: generates the mean group estimates for the Non-OECD subsample (subtracting full sample means) in Table 5 table5ss_oecd: generates the mean group estimates for the OECD subsample (subtracting OECD sample means) in Table 5 table5ss_nonoecd: generates the mean group estimates for the Non-OECD subsample (subtracting Non-OECD sample means) in Table 5 table6fs_all: generates the mean group estimates, allowing for cross-sectional dependence, for the full sample in Table 6 table6fs_oecd: generates the mean group estimates, allowing for cross-sectional dependence, for the OECD subsample (with full sample means) in Table 6 table6fs_nonoecd: generates the mean group estimates, allowing for cross-sectional dependence, for the Non-OECD subsample (with full sample means) in Table 6 table6ss_oecd: generates the mean group estimates, allowing for cross-sectional dependence, for the OECD subsample (with OECD means) in Table 6 table6ss_nonoecd: generates the mean group estimates, allowing for cross-sectional dependence, for the Non-OECD subsample (with Non-OECD means) in Table 6 The dataset descriptions are as follows: pwt61&wdi_75.txt: This dataset includes all the original variables for the 75 countries, used in table2.do, table3.do, table4.do and all table6xx.do files. The variable names are: lrgdpw: log gdp per worker grgdpw: growth rate of gdp per worker lki: log investment share dlki: change in the log investment share linfl: log of 1 plus the inflation rate ltrade: log of exports plus imports as a share of GDP lgov: log of government spending as a share of GDP totgrowth: terms of trade growth lprcredit: log of private credit over GDP lrelprinv: log of relative price of investment xyear: year ccode: country name (string) country: country name (numeric) ccountryxx: country dummy (xx=1-75) pwt6.1_devns75.txt: This dataset includes the variables in deviations away from their full-sample means for the 75 countries. The variable names have "v"s in front to denote deviations. This dataset is used in table1a.do, table1b.do, table5fs_oecd.do and table5fs_nonoecd.do. deviations_nonoecd55.txt: This dataset includes the variables in deviations away from their Non-OECD means for the 55 Non-OECD countries. This dataset is used in table5ss_nonoecd.do. deviations_oecd.txt: This dataset includes the variables in deviations away from their OECD means for the 20 OECD countries. This dataset is used in table5ss_oecd.do. cointg_71.txt: This dataset includes the backward-sum and log GDP for the balanced sample of 71 countries. This dataset is used in the cointegration test in table1c.do.