Abstract

Families of distributions are commonly used to model insurance claims data that require flexible distributional forms in a satisfactory manner, but the specification problem to assess the goodness-of-fit of the hypothesized model can sometimes be a challenge due to the complexity of the likelihood function of the family of distributions involved. The previous work shows that these specification problems can be attacked by means of semi-parametric tests based on generalized method of moment (GMM) estimators. While the approach can be directly applied to both discrete and continuous families of distributions, the paper focuses on developing a testing strategy within a framework of discrete families of distributions. Both the local power analysis and the approximate slope method demonstrate the excellent performance of these tests. The finite-sample performance of the tests, based on both asymptotic and bootstrap critical values, are also discussed and are compared with established methods that require the complete specification of likelihood functions.

Keyword

Claims data ;  goodness-of-fit ;  model specification

Abstract

The availability of high-frequency financial data has led to substantial improvements in our understanding of financial volatility. Most existing literature focuses on estimating the integrated volatility over a fixed period. This article proposes a non-parametric threshold kernel method to estimate the time-dependent spot volatility and jumps when the underlying price process is governed by Brownian semimartingale with finite activity jumps. The threshold kernel estimator combines the threshold estimation for integrated volatility and the kernel filtering approach for spot volatility when the price process is driven only by diffusions without jumps. The estimator proposed is consistent and asymptotically normal and has the same rate of convergence as the estimator studied by Kristensen (2010) in a setting without jumps. The Monte Carlo simulation study shows that the proposed estimator exhibits excellent performance over a wide range of jump sizes and for different sampling frequencies. An empirical example is given to illustrate the potential applications of the proposed method.

Keyword

High‐frequency data ;  jump ;  kernel estimation ;  threshold ;  time dependent ;  volatility

Abstract

The game option, which is also known as Israel option, is an American option with callable features. The option holder can exercise the option at any time up to maturity. This article studies the pricing behaviors of the path-dependent game option where the payoff of the option depends on the maximum or minimum asset price over the life of the option (i.e., the game option with the lookback feature). We obtain the explicit pricing formula for the perpetual case and provide the integral expression of pricing formula under the finite horizon case. In addition, we derive optimal exercise strategies and continuation regions of options in both floating and fixed strike cases.

Keyword

American options ;  Callable feature ;  Lookback game options ;  Path dependent

Abstract

Financial forecasting is the basis for budgeting activities and estimating future financing needs. Applying machine learning and data mining models to financial forecasting is both effective and efficient. Among different kinds of machine learning models, kernel methods are well accepted since they are more robust and accurate than traditional models, such as neural networks. However, learning from multiple data sources is still one of the main challenges in the financial forecasting area. In this paper, we focus on applying the multiple kernel learning models to the multiple major international stock indexes. Our experiment results indicate that applying multiple kernel learning to the financial forecasting problem suffers from both the short training period problem and non-stationary problem. Therefore we propose a novel multiple kernel learning model to address the challenge by introducing the Gompertz model and considering a non-linear combination of different kernel matrices. The experiment results show that our Gompertz multiple kernel learning model addresses the challenges and achieves better performance than the original multiple kernel learning model and single SVM models.

Keyword

financial forecasting ;  multiple kernel learning ;  non-linear kernel combination

Abstract

This paper explores whether the regulation of price limits in financial markets will result in the autocorrelation of stock return series. The results of Monte Carlo Experiment under different error term distribution hypothesis suggest that such price limit mechanism will result in a positive first-order autocorrelation of return series. The results indicates that some statistics in empirical finance literature for testing random walk or market efficiency, for example, the variance ratio of Lo and MacKinlay(1988), may be biased when the stock price is subject to the price limit. This paper also suggests that the further research on the price limit is necessary when more exchanges adopt such regulations in the world.

Keyword

error term distribution hypothesis ;  financial markets ;  Monte Carlo experiment ;  price limit regulation ;  random walk testing ;  stock price ;  stock return series autocorrelation

Abstract

Negative symptoms commonly seen in chronic schizophrenia are related to prefrontal hypodopaminergia. Dysfunction of the catechol-O-methyltransferase (COMT) gene has long been thought to confer susceptibility to schizophrenia because of its catalytic activity for dopamine degradation. The present study is an attempt to perform a quantitative trait test for genetic association between the COMT gene and negative symptoms in a Chinese population. A total of 290 unrelated individuals with schizophrenia patients were recruited and their symptoms were assessed through the Positive and Negative Syndrome Scale (PANSS). The quantitative trait test was performed by the UNPHASED programme to examine the correlation between the scored negative symptoms and the coding single nucleotide polymorphisms (SNPs) present in the COMT gene. The rs4633-rs4680 haplotype showed significant association with the overall score of negative symptoms, and four individual negative symptoms, including blunted affect, emotional withdrawal, poor rapport and passive/apathetic social withdrawal. SNP rs4680 (Val/Met) showed significant association with blunted affect. The present finding suggests that the COMT gene may a etiologically contribute to the severity of negative symptoms in schizophrenia, but its precise mechanism needs further investigating.

Keyword

Catechol-O-methyltransferase (COMT) ;  Chronic schizophrenia ;  Negative symptoms ;  Quantitative trait test

Abstract

We present a semi-parametric method for testing mixing distributions in the mixed Poisson model. The proposed method, which is based on the generalized method of moments, does not demand the complete specification of the probability function but only requires a specification of a set of moment conditions which the model should satisfy. We demonstrate that an explicit expression for moment relations between the mixing and the mixed distributions provides a natural way in selecting moment restrictions and model parameterization. The Monte Carlo evidence suggests that the test has satisfactory performance for moderate size samples.

Keyword

Mixed Poisson distribution ;  Model specification ;  Monte Carlo simulation ;  Semi-parametric method

Abstract

The topic of serial correlation in regression models has attracted a great deal of research in the last 50 years. Most of these studies have assumed that the structure of the error covariance matrix Ω was known or could be consistently estimated from the data. In this article, we describe a new procedure for generating forecasts for regression models with serial correlation based on ordinary least squares and on an approximate representation of the form of the autocorrelation. We prove that the predictors from this specification are asymtotically efficient under some regularity conditions. In addition, we show that there is not much to be gained in trying to identify the correct form of the serial correlation since efficient forecasts can be generated using autoregressive approximations of the autocorrelation. A large simulation study is also used to compare the finite sample predictive efficiencies of this new estimator vis-à-vis estimators based on ordinary least squares and generalized least squares.

Keyword

Abstract

Keyword

Abstract

We analyze, by simulation, the finite-sample properties of goodness-of-fit tests based on residual autocorrelation coefficients (simple and partial) obtained using different estimators frequently used in the analysis of autoregressive moving-average time-series models. The estimators considered are unconditional least squares, maximum likelihood and conditional least squares. The results suggest that although the tests based on these estimators are asymptotically equivalent for particular models and parameter values, their sampling properties for samples of the size commonly found in economic applications can differ substantially, because of differences in both finite-sample estimation efficiencies and residual regeneration methods.

Keyword

Autoregressive-moving average model ;  conditional least squares ;  goodness-of-fit test ;  maximum likelihood ;  partial autocorrelation ;  residual autocorrelation ;  unconditional least squares