In marketing, conjoint analysis a market research technique to understand consumer preferences and evaluate the relative importance of different attributes when choosing a product or service. By dissecting the decision making process, this technique can guide product design, pricing strategies, and market segmentation. Often, multiple linear regression is used to model conjoint data, where the target variable presents the preference or utility score assigned to a product profile, and the explanatory variables are the attributes and levels that define the product profile. The card data shows the various profiles of a product with customer preference scores. Treat the explanatory variables as categorical variable and use multiple linear regression to measure the effect of each attribute. Explain what combination is the most attractive for customers. Guide: In this context, price is separated into 3 tiers, so treat it also as a categorical variable. During encoding, you should create k-1 variables for k distinct categories in each attribute.

Card	Brand	Color	Price	Preference	Brand A	Brand B	Brand C	Color bLACK	Color White
1	A	Black	50	5	1	0	0	1	0
2	A	Black	100	5	1	0	0	1	0
3	A	Black	150	0	1	0	0	1	0
4	A	White	50	8	1	0	0	0	1
5	A	White	100	5	1	0	0	0	1
6	A	White	150	2	1	0	0	0	1
7	B	Black	50	7	0	1	0	1	0
8	B	Black	100	5	0	1	0	1	0
9	B	Black	150	3	0	1	0	1	0
10	B	White	50	9	0	1	0	0	1
11	B	White	100	6	0	1	0	0	1
12	B	White	150	5	0	1	0	0	1
13	C	Black	50	10	0	0	1	1	0
14	C	Black	100	7	0	0	1	1	0
15	C	Black	150	5	0	0	1	1	0
16	C	White	50	9	0	0	1	0	1
17	C	White	100	7	0	0	1	0	1
18	C	White	150	6	0	0	1	0	1