Conditional liquidity shock
factor.formula
The liquidity shock of stock i at time t can be expressed as the residual term of the following time series model:
in:
- :
Amihud's illiquidity index of stock i at time t. The larger the index, the worse the stock liquidity.
- :
The constant term of the illiquidity indicator time series of stock i represents the long-term average level of illiquidity of the stock.
- :
The first-order autoregression coefficient of the illiquidity indicator time series of stock i measures the impact of the illiquidity indicator of the previous period on the illiquidity indicator of the current period.
- :
The residual term of stock i at time t-1 is the liquidity shock value of the previous period.
- :
The first-order moving average coefficient of the illiquidity indicator time series of stock i measures the impact of the liquidity shock in the previous period on the illiquidity indicator in the current period.
- :
The liquidity shock of stock i at time t is also the residual term of the time series model. The larger the value, the worse the current liquidity is than expected.
factor.explanation
The conditional liquidity shock factor predicts the liquidity level of stocks by constructing a time series model (usually ARMA(1,1)), and measures liquidity shocks with the difference between actual liquidity and predicted liquidity (i.e., the residual term). A positive shock indicates that stock liquidity is worse than expected, and a negative shock indicates that stock liquidity is better than expected. This factor captures the risk of sudden changes in short-term liquidity in the market and is usually negatively correlated with future stock returns, indicating that when liquidity suddenly deteriorates, future stock returns are expected to decline. The advantage of this factor is that it not only takes into account the level of liquidity, but also the dynamic changes and unpredictable components of liquidity, thereby more effectively predicting stock returns.