Patrick W. Saart (Wongsaart)

Biography

Publications

• Date
• Type
• Selected

Teaching

Teaching Experience:

Cardiff University, United Kingdom

• BST161 Principles of Finance

• BST178 Topics in Advanced Econometrics: Spatial Econometrics

Newcastle University, United Kingdom

• NBS8257 Applied Econometrics (Semester 2: 2015)

• NBS818 Time Series Analysis (Semester 2: 2015)

• NBS8186 Introductory Econometrics (Semester 1: 2015)

University of Canterbury, New Zealand

• ECON321 Mathematical Techniques in Microeconomics (Semester 1: 2014)

• ECON323 Time Series Methods (Semester 2: 2012, 2013, 2014)

• FINC203 Financial Markets, Institutions and Instruments (Semester 2: 2012, 2013)

• STAT201/FORE222 Applied Statistics, (Semester 1: 2012, 2013, 2014)

• STAT213 Statistical Inferences (Semester 1: 2012, 2013, 2014)

• STAT317/STAT425 Time Series Methods (Semester 2: 2012, 2013, 2014)

• STAT470 Advanced Time Series (Financial Econometrics), (Semester 2: 2013, 2014)

• EMTH119 Engineering Mathematics (Semester 2: 2013, 2014)

• MATH407 Mathematical Finance (Semester 2: 2012)\

My research focuses on developing novel analytical techniques that can help to deal with various types of new data, particularly to find unseen patterns, derive meaningful information, and to make decisions.

My recent projects focuses on:

Functional Time Series: In this project, I first established a new approach to studying temporal dependence in a time series of stochastic processes. Such a method is innovative because it can reduce problems in functional space to simply multivariate time series analysis, which is simpler to deal with in practice. Though this method is applicable to problems in any areas of science, I illustrated the usefulness of the approach in the context of empirical finance. I shew that the resulting model encompassed as special cases dozens of existing models, which had dominated the literature for several decades.

Land-Use Modeling and Prediction in the UK: This project focused on establishment of new empirical methods for analysing agricultural land-use in the UK. The agricultural land-use has far-reaching effects on the environment, e.g. bio-diversity. Therefore, it is imperative that we have a good understanding of famers’ decision making process about the land-use. My project consisted of two important components. Firstly, I constructed an innovative method for analysing high-resolution panel data. Unlike existing methods, my model could (1) handle a censored data problem (i.e. when the dependent variable is censored, values in a certain range are all reported as a single value), and (2) take into consideration possible spatial-dependence caused possibly by some unobserved physical environment of land. I illustrated the superiority of my method in predicting agricultural land-use compared to those currently used by DEFRA. Secondly, I established a set of empirical evidence, which can be used in support of policy-making strategies suggested to DEFRA by Ian Bateman (University of Exeter) in his report, “Public Funding for Public Goods: A Post-Brexit Perspective on Principles for Agricultural Policy”.

Endogeneity Problem and Shape Invariance: In econometrics, endogeneity broadly refers to situations in which an explanatory variable is correlated with the error term. This obstructs successful applications of regression methods. In this project, I established a novel approach for addressing this problem in semiparametric regression.

Working Papers in the Process of Publication:

(Electronic copies of these papers are available in my personal website.)

• Modeling and predicting agricultural land use in England based on spatially high-resolution data (with I. Bateman and N. H. Kim)
• Estimation and Hypothesis Tests of Varying Coefficient Panel Data Model with Spatially Correlated Error Components (with P. M. Robinson and N. H. Kim) (Presented at Econometric Society/Bocconi University Virtual World Congress 2020)
• Understanding Spatial Heterogeneity in Great Britain Agricultural Land-use for Improved Policy Targeting (with I. Bateman and N. H. Kim)
• Commn Factors in Idiosyncratic Volatility and Application in Volatility Forecasting (with Y. Xia) (Presented at Asian Meeting of Econometric Society )

Recent Funding Sucessesses:

Project: Learning Tools for Land Use Analysis and Decision Support Utilising Earth Observation, Natural Capital and Economic Data

Grant Proposal Submitted for: Turing-ONS-HSBC Economic Data Science Award 2018 [Awarded Value £132,000; in collaboration with I. Bateman, J. Davidson, N. H. Kim (U of Exeter), Y. Xia (NUS) and C. Fezzi (Trento)]

Summary: The core of this project is to develop the ability to quantitatively explain the complex interplay between scientific, economic, demographic and statistic factors that are crucial determinants of land-use in the UK. Being able to acquire a better understanding of the processes that drive land-use is essential since land-use is an important factor that determines a wide range of values (e.g. food production, energy production and forestry, water quality and quantity, flooding risk, greenhouse gas emission and storage, recreation and related physical, and mental health) and a number of important environmental concerns (e.g. the long-term decline in biodiversity in the UK). Clearly, the process that determines land-use is exceptionally complex. It depends not only on scientific factors (e.g. climate change), but also on economic determinants (e.g. agricultural price change) and demographic variables (e.g. population growth), to name a few. Furthermore, there are other determinants, which can collectively be described as statistical factors, for example spatial and temporal dependence, and co-movements, (e.g. common trend). The proposed research will conduct an empirical analysis of the interplay between land-use and its important determinants, especially climate change and the agricultural price change, and to construct accurate forecasts of land-use in the UK”.

Please see more information at:

• http://www.exeter.ac.uk/leep/newsandevents/news/articles/economicsdepartmentandleep.php

• https://www.turing.ac.uk/news/turing-hsbc-and-ons-collaborate-economic-data-science-awards

Supervision

I am interested in supervising PhD students in the areas of econometrics and applied statistics. Especially topics related to

• Climate change
• Land-use and the environment
• Financial market and institutuions
• Public health