Table 7 Out-of-sample switching strategy performance of probit model
From: Return direction forecasting: a conditional autoregressive shape model with beta density
| Â | \(r_{t-1}\) | \(r_{t-1}\)+\(tbl_{t-1}\) | \(r_{t-1}\)+\(infl_{t-1}\) | \(r_{t-1}\)+\(svar_{t-1}\) |
|---|---|---|---|---|
Panel A: whole sample | ||||
SR(%) | 6.15 | 6.67 | 6.79 | 3.49 |
\(\Delta\)(U) | 1.69 | 1.92 | 1.92 | 0.64 |
Panel B: Expansion | ||||
SR(%) | 6.27 | 7.82 | 7.76 | 5.40 |
\(\Delta\)(U) | \(-1.47\) | \(-0.98\) | \(-0.99\) | \(-1.83\) |
Panel C: Recession | ||||
SR(%) | 5.63 | \(-0.46\) | 1.77 | \(-5.84\) |
\(\Delta\)(U) | 21.24 | 19.85 | 19.93 | 15.92 |