Timing the market: the economic value of price extremes

Financial Innovation

Table 12 Monthly Regression with Business-cycle Related Variables Controlled

PML _t ^F	BM _t	BM _t+1	PML _t ^F	DE _t	DE _t+1	PML _t ^F	DF Y _t	DFY _t+1	PML _t ^F	DP _t	DP _t+1
0.136^***
0.133^***	0.247^*		0.137^***	0.168		0.135^***	31.118		0.135^***	0.118
0.127^***	19.746^***	-19.627^***	0.134^***	-0.262	0.436	0.130^***	6.632	27.265	0.153^***	17.205^***	-17.207^***
PML _t ^F	DY _t	DY _t+1	PML _t ^F	EP _t	EP+1	P ML _t ^F	INFL _t	INFL _t+1	P ML _t ^F	LT R _t	LTR _t+1
0.139^***	0.115		0.136^***	0.035		0.138^***	-8.621		0.153^***	5.133^***
0.135^***	-0.007	0.123	0.100^***	8.748^***	-8.811^***	0.139^***	-9.618	1.608	0.144^***	4.969^***	2.983^**
PML _t ^F	LT Y _t	LTY _t+1	PML _t ^F	NTIS _t	NTIS _t+1	PML _t ^F	SVAR _t	SVAR _t+1	PML _t ^F	T BL _t	TBL _t+1
0.134^***	1.642		0.136^***	-4.308^**		0.112^***	19.485^**		0.133^***	1.173
0.123^***	43.153^***	-41.711^***	0.130^***	18.680^*	-23.424^**	0.118^***	25.238^**	-13.071	0.134^***	37.059^***	-36.196^***
PML _t ^F	TMS _t	TMS _t+1
0.136^***	0.722
0.142^***	-16.999^*	18.533^**

Note. Our benchmark model is PMG^F_t+1=C+PML_t^F+ε_t+1, where PMG^F and PML^F are filtered observations. Filtered observations are used to alleviate the contamination of autocorrelations in PMG and PML. Regression with business-cycle related variables controlled is presented as follows,
\( {\displaystyle \begin{array}{l}{{PMG^F}_t}_{+1}=C+\alpha {PML_t}^F+{\upbeta}_1{M}_t+{\upvarepsilon}_{\mathrm{t}+1,}\\ {}{{PMG_t}^F}_{+1}=C+{PML_t}^F+{\upbeta}_1{M}_t+{\upbeta}_2{M}_{t+1}+{\upvarepsilon}_{\mathrm{t}+1},\end{array}} \)
where M_t represents business-cycle related variable. The constant C is not reported in the table for space-saving. ***, **, * mean respectively significance at the level of 1%, 5% and 10%