Abstract:This paper proposes the cross-quantilogram to measure the quantile dependence between two time series. We apply it to test the hypothesis that one time series has no directional predictability to another time series. We establish the asymptotic distribution of the cross quantilogram and the corresponding test statistic. The limiting distributions depend on nuisance parameters. To construct consistent confidence intervals we employ the stationary bootstrap procedure; we show the consistency of this bootstrap. Also, we consider the self-normalized approach, which is shown to be asymptotically pivotal under the null hypothesis of no predictability. We provide simulation studies and two empirical applications. First, we use the cross-quantilogram to detect predictability from stock variance to excess stock return. Compared to existing tools used in the literature of stock return predictability, our method provides a more complete relationship between a predictor and stock return. Second, we investigate the systemic risk of individual financial institutions, such as JP Morgan Chase, Goldman Sachs and AIG. This article has supplementary materials online.
What problem does this paper attempt to address?
This paper aims to solve the problem of measuring the quantitative dependence relationship between time series and test whether one time series has directional predictability for another time series. Specifically, the paper proposes the tool of Cross - Quantilogram, which is used to measure the dependence relationship between two time series at different quantile levels and test whether one time series has the predictive ability for another time series at a certain future time point.
### Background and Problems of the Paper
1. **Quantitative Dependence Measurement**:
- Existing methods such as Correlogram and distribution - based statistics (such as variance ratio) perform poorly when dealing with time series with heavy - tailed distributions. Financial time series usually have heavy - tailed characteristics, so a more effective tool is needed to measure this dependence.
- The Cross - Quantilogram measures the dependence relationship between two time series by considering "Quantile Hits", which makes it more effective when dealing with data with heavy - tailed distributions.
2. **Directional Predictability Test**:
- Traditional predictive test methods are mainly based on sign or rank statistics, but these methods often can only provide limited information. The Cross - Quantilogram can provide a more comprehensive description of the relationship, not limited to specific quantile levels.
- The paper proposes a hypothesis - testing method based on the Cross - Quantilogram to test whether one time series has the predictive ability for another time series at a certain future time point.
### Main Contributions
1. **Definition and Properties of Cross - Quantilogram**:
- Defines the Cross - Quantilogram, which is used to measure the dependence relationship between two time series at different quantile levels.
- Derives the asymptotic distribution of the Cross - Quantilogram and its corresponding test statistic.
2. **Construction of Confidence Intervals**:
- In order to construct consistent confidence intervals, the paper adopts the Stationary Bootstrap and proves its consistency.
- Proposes a self - standardization method, which has an asymptotic pivotal distribution under the null hypothesis without estimating the long - term variance or performing bootstrapping.
3. **Extension in Multivariate Settings**:
- Extends the method to multivariate settings and clearly considers the application of the Cross - Quantilogram.
- Applies the Cross - Quantilogram to study the predictability of stock returns and systemic risk.
4. **Use of Conditional Quantiles**:
- Allows the Cross - Quantilogram to be based on Conditional Quantiles, so as to measure the directional dependence relationship between two time series after controlling the information in prediction.
### Empirical Applications
1. **Stock Return Prediction**:
- Uses the Cross - Quantilogram to detect the predictive ability of stock volatility for excess stock returns.
- Compared with the tools in the existing literature, the Cross - Quantilogram provides a more complete description of the relationship.
2. **Systemic Risk Assessment**:
- Studies the systemic risks of individual financial institutions (such as JPMorgan Chase, Morgan Stanley and AIG).
### Conclusion
By proposing the Cross - Quantilogram, the paper provides a new tool to measure the quantitative dependence relationship between time series and test the directional predictability. This method performs well when dealing with data with heavy - tailed distributions, is suitable for financial time series analysis, and shows its effectiveness in empirical applications.