Abstract:Multivariate Hawkes processes (MHP) are a class of point processes in which events at different coordinates interact through mutual excitation. The weighted adjacency matrix of the MHP encodes the strength of the relations, and shares its support with the causal graph of interactions of the process. We consider the problem of testing for causal relationships across the dimensions of a marked MHP. The null hypothesis is that a joint group of adjacency coefficients are null, corresponding to the absence of interactions. The alternative is that they are positive, and the associated interactions do exist. To this end, we introduce a novel estimation procedure in the context of a large sample of independent event sequences. We construct the associated likelihood ratio test and derive the asymptotic distribution of the test statistic as a mixture of chi squared laws. We offer two applications on financial datasets to illustrate the performance of our method. In the first one, our test reveals a deviation from a static equilibrium in bidders' strategies on retail online auctions. In the second one, we uncover some factors at play in the dynamics of German intraday power prices.
What problem does this paper attempt to address?
The problem that this paper attempts to solve is to test whether there is a causal relationship between different coordinates in Multivariate Hawkes Processes (MHP). Specifically, the paper focuses on how to determine through statistical tests whether certain specific interactions exist in the marked Multivariate Hawkes Processes. The paper proposes a method based on the likelihood - ratio test to detect whether a given set of adjacency coefficients are zero (i.e., no interaction exists), or whether these coefficients are positive (i.e., interaction exists). This problem is particularly important in financial data analysis, such as the analysis of bidder behavior in online auctions and the dynamic study of German intraday electricity prices.
### Main contributions of the paper
1. **Proposing a new estimation procedure**: The paper introduces a new estimation method in the context of a large number of independent event sequences and constructs the corresponding likelihood - ratio test.
2. **Deriving the asymptotic distribution of the test statistic**: The paper proves that under the null hypothesis, the asymptotic distribution of the likelihood - ratio test statistic is a mixture of chi - square distributions ($\chi^{2}$ distributions).
3. **Providing two practical applications**:
- **Application in online auctions**: Through testing, it reveals the deviation of bidder strategies from the static equilibrium state.
- **Application in German intraday electricity prices**: It discovers some factors that affect the electricity price dynamics, especially the role of liquidity constraints.
### Mathematical model and assumptions
The paper uses the linear form of the Multivariate Hawkes Process, and its intensity function is defined as:
\[
\lambda_{k,t}=\mu_{k}\left(\frac{t}{T}\right)+\sum_{l = 1}^{K}\sum_{t_{l}^{i}<t}\varphi_{kl}(t - t_{l}^{i},X_{l}^{i})
\]
where:
- $\mu_{k}$ is the baseline intensity.
- $\varphi_{kl}$ is the kernel function from coordinate $l$ to coordinate $k$, which describes the mutual excitation effect between events.
### Mathematical formulation of the test problem
The paper considers the following hypothesis - testing problems:
- **Null hypothesis**: For all $(k, l)\in J$, $\int_{0}^{\infty}\varphi_{kl}(s)ds = 0$, indicating no interaction.
- **Alternative hypothesis**: For all $(k, l)\in J$, $\int_{0}^{\infty}\varphi_{kl}(s)ds>0$, indicating the existence of interaction.
### Main results
1. **Asymptotic distribution of the likelihood - ratio test**: Under a series of assumption conditions, the paper proves that the asymptotic distribution of the likelihood - ratio test statistic $\Lambda_{n}$ is a mixture of chi - square distributions.
2. **Consistency and asymptotic distribution of the maximum likelihood estimate**: As a by - product of the likelihood - ratio test, the paper also obtains the consistency and asymptotic distribution of the maximum likelihood estimate (MLE).
### Practical applications
The paper demonstrates the effectiveness of the proposed method through two practical examples:
- **Online auctions**: Through testing, it is found that the interaction between bidders significantly affects the arrival time of bids, resulting in the transaction price deviating from the static Vickrey equilibrium.
- **German intraday electricity market**: It studies the price - formation mechanism of hourly electricity futures and finds that in addition to the classic mean - reversion, the price dynamics also exhibit an inertia, which may be due to liquidity limitations.
### Conclusion
This paper provides an effective statistical method for detecting causal relationships in Multivariate Hawkes Processes and verifies the effectiveness of this method through actual data. This method has important application value in financial data analysis.