1. The lower and upper [I(0) and I(1)] non-standard F-statistics are 3.602 (2.688) and 4.787 (3.698) at 1% (5% ) significance level, respectively. LR and SR are short-and long-run. BG_LM is the Breusch Godfrey serial correlation Lagrange Multiplier F-statistics, while RESET is Ramsey Regression Equation Specification Error Test. *, **, *** indicate statistical significance at the 10 percent, 5 percent, and 1 percent significance levels, respectively. Evidence from Table 6 shows rejection of the null hypotheses;Narayan;Notes: For the asymmetric cointegration Wald test, the non-standard critical values were derived by,2005
2. This means that (i) changes in positive and negative sentiment have a short-term causal impact on stock returns, (ii) IISI is a better proxy as a short-term predictor of portfolio returns, but the predictive power is decreasing in portfolio size-capitalization and slightly higher for value than growth portfolio. This evidence suggests that mean reversion in positive and negative sentiments is associated with shortterm mispricing, which arbitrage can more easily correct to dampen further deviation of prices from their fundamental values. Overall, changes in sentiment persistence have significant implications for asset pricing and the predictive power of sentiment on stock returns;Consistent with Schmeling (2009) and Lemmon and Portniaguina,2006
3. Home and foreign investor sentiment and the stock returns;D B Aissia;The Quarterly Review of Economics and Finance,2016
4. Do exchange rate changes improve the trade balance: An asymmetric nonlinear cointegration approach;A C Arize;International Review of Economics & Finance,2017
5. Non-stationarity and nonlinearity in inflation rate: Some further evidence;A C Arize;International Review of Economics Finance,2012