A baseball team's pricing department would like to determine the price to sell each ticket to each game. This description is simplified from its real complexity; if you're an expert in the ticket pricing industry, please do not rely on your expertise to fill in all the extra complexity.

Currently there is a single price for each section's tickets - for example, to sit in Section 110 cost the same amount (75$) at every one of the team's 81 home games. The VP of pricing believes this leads to two kinds of bad outcomes. For some games(for example, a weekday afternoon game against a bad opponent), the seat never gets sold because nobody wants to pay that much, and for other games(for example, a Saturday night game against a very good opponent) all the seats in the sections are sold for $75 but could've been sold for more.

The VP of Pricing has decided to use analytics to determine the price for a ticket in each section at each game, based on the game's characteristics(day, time, opponent, etc.) and the ticket price.

At first, there is very little data, because each section's tickets have been sold at a single price for all 81 games.

a)Select all of the models/approaches the pricing department could use to estimate demand for a section's tickets at a specific price until enough data is collected.

a. A/B testing, with two different prices being set for similar games

b. Bayesian analysis, with an initial baseline distribution for each section/price provided by the VP of Pricing and updated as each game's ticket sales provide a new data point

c.) Linear regression based on past sales data, even though the prices being set are outside the range of prices in the data

d.) Markov chain, with each state being the price of a ticket

e) Support Vector Machine based on past sales data, even though the prices being set are outside the range of the prices in the data.