A Research on the Correlation between Crime Rates and Social Indices

碩士 === 國立清華大學 === 公共政策與管理碩士在職專班 === 106 === ABSTRACT From 1980 to 2016, the continuous growth of crime rates, offender rates and unemployment rates in Taiwan is worrisome. Therefore, We use the latest 18-year panel data of 20 cities in Taiwan to study the correlation between crime rates and social i...

Full description

Bibliographic Details
Main Authors: Lien, Kuo-Feng, 連國豐
Other Authors: Lu, Shu-Shiuan
Format: Others
Language:zh-TW
Published: 2018
Online Access:http://ndltd.ncl.edu.tw/handle/q5y48t
id ndltd-TW-106NTHU5610006
record_format oai_dc
spelling ndltd-TW-106NTHU56100062019-05-16T00:52:41Z http://ndltd.ncl.edu.tw/handle/q5y48t A Research on the Correlation between Crime Rates and Social Indices 犯罪率與社會指標相關性研究 Lien, Kuo-Feng 連國豐 碩士 國立清華大學 公共政策與管理碩士在職專班 106 ABSTRACT From 1980 to 2016, the continuous growth of crime rates, offender rates and unemployment rates in Taiwan is worrisome. Therefore, We use the latest 18-year panel data of 20 cities in Taiwan to study the correlation between crime rates and social indices based on Panel Vector autoregression (PVAR)and then provide the related policy recommendations. From the perspective of Granger causality test, the unemployment rates are not the cause of crime rates but the crime detection rates are the causes of the crime rates. Furthermore, both the crime rates and the crime detection rates cause the changes of the unemployment rates. Next, the crime rates are the causes of the changes of the crime detection rates. However, the unemployment rates do not affect the crime detection rates. Additionally, we find that the Impulse Response of the unemployment rates on the crime rates or that of the crime rates on the unemployment rates is negative at very beginning and turns positive gradually. Yet, this non-simple-linear relationship does not converge to zero. Finally, based on the the variance decomposition analysis, we find that 81% of the crime rates, the unemployment rates and the crime detection rates are influenced by their past experiences. Moreover, the past crime rates explained 17.60% the current unemployment rates, which implies that public security affects unemployment rates significantly. Next, 12.24% of the crime detection rates are explained by the past crime rates. These results suggest that crime detection rates are significantly affected by crime rates. Accordingly, this empirical research supports argument that crime rates can be effectively reduced when the government continue improving the crime detection rates of the police. Lu, Shu-Shiuan 盧姝璇 2018 學位論文 ; thesis 55 zh-TW
collection NDLTD
language zh-TW
format Others
sources NDLTD
description 碩士 === 國立清華大學 === 公共政策與管理碩士在職專班 === 106 === ABSTRACT From 1980 to 2016, the continuous growth of crime rates, offender rates and unemployment rates in Taiwan is worrisome. Therefore, We use the latest 18-year panel data of 20 cities in Taiwan to study the correlation between crime rates and social indices based on Panel Vector autoregression (PVAR)and then provide the related policy recommendations. From the perspective of Granger causality test, the unemployment rates are not the cause of crime rates but the crime detection rates are the causes of the crime rates. Furthermore, both the crime rates and the crime detection rates cause the changes of the unemployment rates. Next, the crime rates are the causes of the changes of the crime detection rates. However, the unemployment rates do not affect the crime detection rates. Additionally, we find that the Impulse Response of the unemployment rates on the crime rates or that of the crime rates on the unemployment rates is negative at very beginning and turns positive gradually. Yet, this non-simple-linear relationship does not converge to zero. Finally, based on the the variance decomposition analysis, we find that 81% of the crime rates, the unemployment rates and the crime detection rates are influenced by their past experiences. Moreover, the past crime rates explained 17.60% the current unemployment rates, which implies that public security affects unemployment rates significantly. Next, 12.24% of the crime detection rates are explained by the past crime rates. These results suggest that crime detection rates are significantly affected by crime rates. Accordingly, this empirical research supports argument that crime rates can be effectively reduced when the government continue improving the crime detection rates of the police.
author2 Lu, Shu-Shiuan
author_facet Lu, Shu-Shiuan
Lien, Kuo-Feng
連國豐
author Lien, Kuo-Feng
連國豐
spellingShingle Lien, Kuo-Feng
連國豐
A Research on the Correlation between Crime Rates and Social Indices
author_sort Lien, Kuo-Feng
title A Research on the Correlation between Crime Rates and Social Indices
title_short A Research on the Correlation between Crime Rates and Social Indices
title_full A Research on the Correlation between Crime Rates and Social Indices
title_fullStr A Research on the Correlation between Crime Rates and Social Indices
title_full_unstemmed A Research on the Correlation between Crime Rates and Social Indices
title_sort research on the correlation between crime rates and social indices
publishDate 2018
url http://ndltd.ncl.edu.tw/handle/q5y48t
work_keys_str_mv AT lienkuofeng aresearchonthecorrelationbetweencrimeratesandsocialindices
AT liánguófēng aresearchonthecorrelationbetweencrimeratesandsocialindices
AT lienkuofeng fànzuìlǜyǔshèhuìzhǐbiāoxiāngguānxìngyánjiū
AT liánguófēng fànzuìlǜyǔshèhuìzhǐbiāoxiāngguānxìngyánjiū
AT lienkuofeng researchonthecorrelationbetweencrimeratesandsocialindices
AT liánguófēng researchonthecorrelationbetweencrimeratesandsocialindices
_version_ 1719172123094679552