tmichaelross

Maybe heres a better question: Would it be wiser to abandon the undergrad thesis idea and focus instead on nailing down some more math and stats classes during my senior year? Would grad schools even care about an undergrad thesis?

Hello all, I've come looking for help with an undergrad thesis I'm putting together. It is based on the theory of Henry George, that the taxation of rent can support the functions of government. I'm limiting this to the local level, and I plan to estimate the value of all land in my city. This has been done by multiple economists in the past in other cities, so I won't be inventing methods, simply using others'. The method is this: First I must gather all observed sales of vacant lots in the city in the past 2 years, only using those that are qualified. I will reduce these sale prices to a price per square foot. I will then use multivariate regression to determine the influence of several variables on the price per square foot of bare land. These variable will include distance from the central business district, distance from grocery stores, neighborhood quality (physical and social), etc. Here is the problem: I haven't taken econometrics or much statistics, I am very inexperienced when it comes to regression analysis. I have Stata, I know how to run regressions, but I'm not sure how to analyze the output. Further, once I determine how much and with what statistical significance each variable influences the selling price of a square foot of land, I'm not sure how to apply that to every square foot in the city to give me a concrete estimate of total land value. I know this can be done, because it has been done. The problem is that the papers where these orginal methods are described were written by and for professional economists, and not necessarily for novices like me who are not familiar with regression analysis. So if anyone catches the drift of what I'm trying to do, has experience in real estate valuation or regression analysis, I would be very interested in their insight. Thanks, T