Abstract
We jointly evaluate the two directional changes (increase/decrease and acceleration/deceleration) in monthly manufacturing production activity predicted by production managers using a new market-timing test. Besides determining that the forecasts are useful in predicting an increase/decrease and acceleration/deceleration in monthly production activity, joint evaluation of the forecast illustrates the advantage of the new test. Moreover, we find that the results are broadly robust to two data vintages and outlying observations.
Notes
1 See http://www.meti.go.jp/english/statistics/tyo/iip/index.html for detailed information.
2 The eleven industries are as follows: Iron and Steel, Non-Ferrous Metals, Fabricated Metals, General Machinery, Electrical Machinery, Information and Communication Electronics Equipment, Electronic Parts and Devices, Transport Equipment, Chemicals, Pulp and Paper and Others.
3 See Croushore (Citation2011) for detailed discussions in a broad context.
4 See Ash et al. (Citation1998) and Sinclair et al. (Citation2010) for details. In particular, Sinclair et al. (Citation2010) describe and remark on the test statistics in the 4 × 4 case.
5 See Chou and Chu (Citation2011) for remarks.
6 In the 4 × 4 case for the FE test, hypergeometric distribution is used to directly calculate the probability of independence. Thus, only p-values are reported.