| dc.identifier.uri | http://hdl.handle.net/11401/77412 | |
| dc.description.sponsorship | This work is sponsored by the Stony Brook University Graduate School in compliance with the requirements for completion of degree. | en_US |
| dc.format | Monograph | |
| dc.format.medium | Electronic Resource | en_US |
| dc.language.iso | en_US | |
| dc.publisher | The Graduate School, Stony Brook University: Stony Brook, NY. | |
| dc.type | Dissertation | |
| dcterms.abstract | The mixture distribution hypothesis is widely used to explain the behavior of returns and volumes in security trading in single-security settings, and the unobservable information arrival process in models based on this hypothesis is the key factor that determines the conditional distributions of returns and volumes. In order to investigate whether the information arrival processes of different securities may interact with each other, in the first part of this dissertation I extended the framework of mixture distribution hypothesis to a multiple-security setting, in which the unobservable information arrival processes of different securities may potentially interact with each other through vector autoregression. Then I picked 43 large capitalization stocks publicly traded on US exchanges, grouped them into 11 pairs and 7 triplets with the same industrial sectors, and estimated the multi-security mixture distribution model using these data. My estimation results show that contemporary correlations in the shocks to information arrival processes are more common than cross-security historical dependencies in information arrival processes. However, for 9 out of 18 groups of stocks in industries such as banking, retail, consumer goods and telecommunication services, the cross-security historical dependencies are of both statistical and practical significance. Furthermore, such dependencies are asymmetric in the sense that large capitalization stocks tend to give more impacts to small capitalization stocks than in the other way. My estimation method is based on transforming the likelihood maximization problem into an equation-solving problem involving a high-dimensional integral, and then I use Stochastic Approximation and Markov Chain Monte Carlo simulations to search for the equation’s solution. My simulation study shows that point estimates produced by this method are close to the true parameter values, but the estimated confidence intervals may be not wide enough to cover true parameter values with the corresponding probabilities. In the second part of this dissertation I applied the same model using ETF data to investigate spillover effects in information arrivals among international stock portfolios and among different US asset classes. My estimation results show that the information arrival process of the US stock portfolio can heavily impact those of other countries’ stock portfolios. Similarly, the information arrival process of the large-capitalization stock portfolio can heavily impact those of mid- and small-capitalization stock portfolios. However, both of these two kinds of impacts are unidirectional. | |
| dcterms.abstract | The mixture distribution hypothesis is widely used to explain the behavior of returns and volumes in security trading in single-security settings, and the unobservable information arrival process in models based on this hypothesis is the key factor that determines the conditional distributions of returns and volumes. In order to investigate whether the information arrival processes of different securities may interact with each other, in the first part of this dissertation I extended the framework of mixture distribution hypothesis to a multiple-security setting, in which the unobservable information arrival processes of different securities may potentially interact with each other through vector autoregression. Then I picked 43 large capitalization stocks publicly traded on US exchanges, grouped them into 11 pairs and 7 triplets with the same industrial sectors, and estimated the multi-security mixture distribution model using these data. My estimation results show that contemporary correlations in the shocks to information arrival processes are more common than cross-security historical dependencies in information arrival processes. However, for 9 out of 18 groups of stocks in industries such as banking, retail, consumer goods and telecommunication services, the cross-security historical dependencies are of both statistical and practical significance. Furthermore, such dependencies are asymmetric in the sense that large capitalization stocks tend to give more impacts to small capitalization stocks than in the other way. My estimation method is based on transforming the likelihood maximization problem into an equation-solving problem involving a high-dimensional integral, and then I use Stochastic Approximation and Markov Chain Monte Carlo simulations to search for the equation’s solution. My simulation study shows that point estimates produced by this method are close to the true parameter values, but the estimated confidence intervals may be not wide enough to cover true parameter values with the corresponding probabilities. In the second part of this dissertation I applied the same model using ETF data to investigate spillover effects in information arrivals among international stock portfolios and among different US asset classes. My estimation results show that the information arrival process of the US stock portfolio can heavily impact those of other countries’ stock portfolios. Similarly, the information arrival process of the large-capitalization stock portfolio can heavily impact those of mid- and small-capitalization stock portfolios. However, both of these two kinds of impacts are unidirectional. | |
| dcterms.available | 2017-09-20T16:52:39Z | |
| dcterms.contributor | Conesa, Juan C | en_US |
| dcterms.contributor | Brusco, Sandro | en_US |
| dcterms.contributor | Zhao, Jake | en_US |
| dcterms.contributor | Xing, Haipeng. | en_US |
| dcterms.creator | Wang, Zhenning | |
| dcterms.dateAccepted | 2017-09-20T16:52:39Z | |
| dcterms.dateSubmitted | 2017-09-20T16:52:39Z | |
| dcterms.description | Department of Economics | en_US |
| dcterms.extent | 174 pg. | en_US |
| dcterms.format | Application/PDF | en_US |
| dcterms.format | Monograph | |
| dcterms.identifier | http://hdl.handle.net/11401/77412 | |
| dcterms.issued | 2016-12-01 | |
| dcterms.language | en_US | |
| dcterms.provenance | Made available in DSpace on 2017-09-20T16:52:39Z (GMT). No. of bitstreams: 1
Wang_grad.sunysb_0771E_13029.pdf: 5265764 bytes, checksum: 1fae087aaf1f50e909b0c437cfdb0495 (MD5)
Previous issue date: 1 | en |
| dcterms.publisher | The Graduate School, Stony Brook University: Stony Brook, NY. | |
| dcterms.subject | Finance | |
| dcterms.subject | Information, Mixture Distribution Hypothesis, Spillover | |
| dcterms.title | Essays on the Spillover Effects of Information Arrivals in Security Trading | |
| dcterms.type | Dissertation | |
