Subset the data set based on the location, day of the week, type of collision, and lighting condition. Compare these subsets of data to find interesting patterns. Can you identify any links between crash fatality and the aforementioned variables? Are there any missing values? Which strategy should you use to handle the missing values? Because many of the variables are categorical, you should consider transforming them into dummy variables prior to the analysis.

ID	County	City	Weekday	Severity	ViolCat	ClearWeather	Month	CrashType	Highway
1	SAN DIEGO	SAN DIEGO	7	1	8	0	1	A	0
2	HUMBOLDT	UNINCORPORATED	4	1	8	1	1	A	1
3	VENTURA	OXNARD	2	1	12	1	2	A	0
4	STANISLAUS	UNINCORPORATED	4	1	1	1	1	A	0
5	MENDOCINO	UNINCORPORATED	5	1	1	1	1	A	1
6	LOS ANGELES	LONG BEACH	7	1	3	1	3	A	0
7	LOS ANGELES	LOS ANGELES	4	1	3	0	3	A	0
8	CALAVERAS	UNINCORPORATED	1	1	1	1	2	A	1
9	SAN BERNARDINO	HESPERIA	2	1	1	1	1	A	0
10	VENTURA	OXNARD	5	0	8	1	1	A	0
11	VENTURA	OXNARD	6	0	8	0	1	A	0
12	ORANGE	FULLERTON	4	0	9	1	2	A	0
13	SAN DIEGO	CHULA VISTA	1	0	3	1	1	A	0
14	ALAMEDA	OAKLAND	6	0	1	1	1	A	0
15	LOS ANGELES	LOS ANGELES	5	0	9	1	3	A	0
16	SANTA CLARA	MORGAN HILL	4	0	8	1	3	A	0
17	LOS ANGELES	LOS ANGELES	3	0	9	1	3	A	0
18	SAN JOAQUIN	UNINCORPORATED	3	0	8	1	3	A	0
19	LOS ANGELES	LOS ANGELES	5	0	9	1	4	A	0