因為生活方式的改變和國內旅遊的人數增加，交通工具的需求大增汽車密度也逐漸攀升，交通事件成了政府必須解決並改善的問題。本研究希望在美國馬里蘭州蒙哥馬利郡交通違規事件資料庫中分析各種交通違規事件，並用核密度估計方法評估局部機率事件的發生進而評估整體區域，將一組不連續的點資料集，利用函數呈現成連續型的型態，可以輕鬆地將資料呈現在視覺資料上提升資料的可用性，藉此找出減少違規事件以及車禍發生機率的方法。
我們觀察到最明顯的現象為沒繫安全帶與死亡，或與違規造成事故有顯著關係；酒駕跟其他事件則皆有顯著關係；七個事件的強度函數圖觀察到違規事件幾乎都發生在市中心且人口密集的主要道路上。

Due to lifestyle changes and the number of people traveling in the country increase, the demand for transportation and the density of vehicles have increased. Traffic violations have become a problem that the government must solve and improve. This study analyzes various traffic violations in the data of traffic violations in Montgomery County, Maryland, USA. Using the kernel density estimation method to estimate the occurrence of the local probability and then evaluate the overall area. Kernel estimation can make a set of discontinuous point data into continuous estimation surface. Therefore, we can easily demonstrate the data by plots and try to find solutions to reduce the incidence of violations and car accidents.
The most obvious phenomenon we observed was that there was significant relationship between the absence of no seat belts with deaths, or with the violation of regulations. Drunk driving has a significant relationship with other events. From the intensity function plots for the seven events, we observe that violations often occur on densely populated main roads.

目錄
第一章 緒論...................................1
第二章 研究方法...............................4
第一節 強度函數 (Intensity Function)...........4
第二節 卜瓦松分配 (Poisson Distribution).......5
第三節 卜瓦松過程檢定方法......................7
第四節 相關性檢定.............................8
第三章 案例分析...............................9
第一節 違規資料敘述...........................10
第二節 違規分析結果...........................16
第四章 結論..................................32
參考文獻.....................................34

圖目錄
3.1 Ripley’s K-function 在均勻假設下 (I)..........20
3.2 Ripley’s K-function 在均勻假設下 (II).........21
3.3 Ripley’s K-function 在不均勻假設下 (I)........22
3.4 Ripley’s K-function 在不均勻假設下 (II).......23
3.5 Cross-type K-function(I).....................24
3.6 Cross-type K-function(II)....................25
3.7 Cross-type K-function(III)...................26
3.8 Cross-type K-function(IV)....................27
3.9 事件分布及其強度圖 (I)........................28
3.10 事件分布及其強度圖 (II)......................29
3.11 事件分布及其強度圖 (III).....................30
3.12 事件分布及其強度圖 (IV)......................31

表目錄
3.1 二元類別變數次數分配表.........................10
3.2 二元類別變數次數分配表 (百分比).................11
3.3 性別違規次數分配表.............................11
3.4 性別違規次數分配表 (百分比).....................11
3.5 安全帶違規 vs. 人身傷害........................12
3.6 安全帶違規 vs. 人身傷害 (百分比)................12
3.7 安全帶違規 vs. 財產損害.........................12
3.8 安全帶違規 vs. 財產損害 (百分比).................12
3.9 安全帶違規 vs. 死亡.............................12
3.10 安全帶違規 vs. 死亡 (百分比)....................12
3.11 安全帶違規 vs. 因違規造成事故...................13
3.12 安全帶違規 vs. 因違規造成事故 (百分比)...........13
3.13 酒駕 vs. 安全帶違規............................13
3.14 酒駕 vs. 安全帶違規 (百分比)....................13
3.15 酒駕 vs. 人身傷害..............................14
3.16 酒駕 vs. 人身傷害 (百分比)......................14
3.17 酒駕 vs. 財產損害...............................14
3.18 酒駕 vs. 財產損害 (百分比).......................14
3.19 酒駕 vs. 因違規造成事故..........................15
3.20 酒駕 vs. 因違規造成事故 (百分比)..................15
3.21 無安全帶違規情況下人身傷害 vs. 死亡................16
3.22 無安全帶違規情況下人身傷害 vs. 死亡 (百分比)........16
3.23 有安全帶違規情況下人身傷害 vs. 死亡................16
3.24 有安全帶違規情況下人身傷害 vs. 死亡 (百分比)........16
3.25 Quadrat Counting Test 結果表........................17
3.26 Kolmogorov-Smirnov Test 結果表.....................18
3.27 Ripley’s K-function 結果表.........................19

[1] 王政雲. 犯罪點資料之空間分布的正規化估計及關聯性分析. PhD thesis, 國立清華大學, 2016.
[2] A.Baddeley, E.Rubak, and R.Turner. Spatial point patterns : methodology and applications with R. Chapman and Hall/CRC, 2015.
[3] T.C.Bailey and A.C.Gatrell. Interactive spatial data analysis, volume 413.1995.
[4] W.R.Black. Highway accidents : a spatial and temporal analysis. Transportation Research Record, 1318:75–82,1991.
[5] W.R.Black. Network auto correlation in transport network and flow systems. Geographical Analysis, 24(3):207–222,1992.
[6] G.Borruso. Network density estimation : analysis of point patterns over a network. In International Conference on Computational Science and Its Applications, pages
126–132. Springer, 2005.
[7] Y.-M.Chang, N.-J.Hsu, andH.-C.Huang. Semiparametric estimation and selection for nonstationary spatial covariance functions. Journal of Conputational and Graphical Statistics, 19(1):117–139,2010.
[8] R.Factor. The effect of traffic tickets on road traffic crashes. Accident Analysis& Prevention, 64:86–91,2014.
[9] B.Flahaut, M.Mouchart, E.SanMartin, and I.Thomas.The local spatial auto correlation and the kernel method for identifying black zones : A comparative approach. Accident Analysis & Prevention, 35(6):991–1004,2003.
[10] S.Hashimoto, S.Yoshiki, R.Saeki, Y.Mimura, R.Ando, and S.Nanba. Development and application of traffic accident density estimation models using kernel density estimation. Journal of traffic and transportation engineering (English edition), 3(3):262–270,2016.
[11] R. K.Jones. Identification of general risk-management countermeasures for unsafe driving 
actions. volume iii: a definitional study of speeding, following too closely, and driving left of center. finalreport. 1981.
[12] J.E.Kelsall and P.J.Diggle. Non-parametric estimation of spatial variation in relative risk. Statistics in medicine, 14(21-22):2335–2342,1995.
[13] J.E.Kelsall and P.J.Diggle. Spatial variation in risk of disease : a nonparametric binary regression approach. Journal of the Royal Statistical Society : Series C
(Applied Statistics), 47(4):559–573,1998.
[14] N.Levine. The location of late night bars and alcohol-related crashes in houston, texas. Accident Analysis & Prevention, 107:152–163,2017.
[15] N.Levine, K.E.Kim, and L.H.Nitz. Spatial analysis of honolulu motor vehicle crashes: I.spatial patterns. Accident Analysis & Prevention, 27(5):663–674,1995.
[16] D.Richards. Relationship between speed and risk of fatal injury : pedestrians and car occupants.2010.
[17] F.Sagberg and R.Ingebrigtsen. Effects of a penalty point system on traffic violations. Accident Analysis & Prevention, 110:71–77,2018.
[18] Z.Xie and J.Yan. Kernel density estimation of traffic accidents in a network space. Computers, environment and urban systems, 32(5):396–406,2008.
[19] I.Yamada and J.-C.Thill. Comparison of planar and network k-functions in traffic accident analysis. Journal of Transport Geography, 12(2):149–158,2004.
[20] Y.Zheng and J.M.Phillips. l∞ 
error and bandwidth selection for kernel density estimates of large data. In Proceedings of the 21th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pages 1533–1542. ACM, 2015.