STATISTICS FOR BUSINESS AND ECONOMICS

QUESTION 1 This data is from the following research article on tuna sales in Japan. BFT stands for Bluefin Tuna. FQ stands for the fresh quantity- the number of tons of fresh (not frozen) tuna exchanged in the market in the given month. FP stands for the fresh price. D stands for domestic (which in this context means in Japan) while I is international. * * * * * I dont know squat about the tuna harvest but the article tells me the quantities are in metric tons. A Z-score is a great way to summarize data for variables that we dont really have a feel for. If someone tells you that some number of tons were hauled in by sellers this month that probably doesnt mean much to you but if we convert it to Zs we can tell if thats a lot. Use bft_fq_i to find the Z-score for 500 metric tons. 2 points

QUESTION 2 This data is from the following research article on tuna sales in Japan. BFT stands for Bluefin Tuna. FQ stands for the fresh quantity- the number of tons of fresh (not frozen) tuna exchanged in the market in the given month. FP stands for the fresh price. D stands for domestic (which in this context means in Japan) while I is international. * * * * * Unfortunately prices are in Yen per kilogram. There are about 100 yen per dollar and 2.2 pounds per kilogram so to combine we should divide each price by 220. The original price is bft_fp_d; lets rename the converted price price_jp. What command lets us sum up a variable by values of another variable? Find the average of price_jp in June. (Meaning the average price in all 14 Junes- use the function from last week to crank this out.) 2 points

QUESTION 3 There are two variables in the data that contain prices: bft_fp_d and bft_fp_i. We converted the first one into price_jp so lets make the second one into price_int again converting by dividing the price by 220. Create a new variable called price_combined that is the “average” of these two variables. (Just add them and divide it by two. Yes a weighted average would be better but we are keeping things really simple.) What is the IQR of this new variable? 2 points

QUESTION 4 There are two variables in the data that contain quantities: bft_fq_d and bft_fq_i. Create a new variable called quant_total that is the sum of these two variables. What is the median of this new variable? 2 points

QUESTION 5 This data is from the following research article on tuna sales in Japan. Load the Bluefin Tuna data (Bluefin_short.csv). BFT stands for Bluefin Tuna. FQ stands for the fresh quantity- the number of tons of fresh (not frozen) tuna exchanged in the market in the given month. FP stands for the fresh price. D stands for domestic (which in this context means in Japan) while I is international Find the correlation between quant_total and price_combined in summer. 2 points

QUESTION 6 Calculate the SQUARE ROOT of the following mathematical expression using Stata: 4(99/31)log(5.44)+12 2 points

QUESTION 7 Use the Fragile Families data for the next few problems. What is the overall correlation between the mothers age (mom_age) and her income (mom_income)? Be accurate to within at least 0.01; as always you can enter as many digits past that as you like. 2 points

QUESTION 8 Use the Fragile Families data for the next few problems. The Fragile Familes data was created to investigate the conditions and capabilities of unmarried parents. They did include a sample of married folks to provide grounds for comparison. Limiting the data to the married sample (for whom married==1) what is the correlation between the mothers age (mom_age) and her income (mom_income)? Be accurate to within at least 0.01; as always you can enter as many digits past that as you like. 2 points QUESTION 9 Use the Fragile Families data for the next few problems. What is the median income (mom_income) in this data? Be accurate to within at least 0.01; as always you can enter as many digits past that as you like. 2 points

QUESTION 10 Use the Fragile Families data for the next few problems. Using the median income make a new variable called hi_inc that is 1 if the mothers income is greater or equal to the median and 0 otherwise. What % of the data are in the high income group (i.e. hi_inc is 1)? Enter the precise answer below i.e. if the answer is 12.34% enter 12.34 below. 2 points

QUESTION 11 Use the Fragile Familes data for this problem. Find the 95th percentile of the number of years the mother has lived in her current neighborhood (years_home). Be accurate to within at least 0.01; as always you can enter as many digits past that as you like. 2 points

QUESTION 12 Use the Fragile Familes data for this problem. Limiting the data to the married sample (for whom married==1) calculate the IQR of the number of years the mother has lived in her current neighborhood. Be accurate to within at least 0.01; as always you can enter as many digits past that as you like. 2 points

QUESTION 13 One characteristic of the Fragile Families data is that there are no missing values. For all of nearly 5000 interviews if someone didnt want to ask a question or if it wasnt appropriate (like a question about someone whos partner died) they put in -3 if the information is “missing” and -6 if the question was skipped. Basically any negative value indicates missing data. To fix that we will replace negative numbers with the “.” character as follows: replace years_home = . if years_home<0 Now that you have removed those observations from consideration find again the 95th percentile of the years_home variable. 3 points

QUESTION 14 Make sure you have replaced the missing values. as asked in the last question. Limiting the data to the married sample (for whom married==1) calculate the IQR of the number of years the mother has lived in her current neighborhood. Be accurate to within at least 0.01; as always you can enter as many digits past that as you like. 3 points

QUESTION 15 The last two questions showed how marking the missing data as “missing” influences the results of our years_home variable. Lets look at another variable years_father (how long the mother knew the father before she got pregnant). Lets see how those “missing” values screws things up. See how many unique values show up in the years_father variable: remember how to check that? 2 points

QUESTION 16 Like years_home the years_father variable has some negative values. How many observations have negative values? There are a few ways to figure this out: you already know tab for example and that will work but you will have to add the numbers up yourself. Theres also a command: count can be used with if to learn how many observations meet certain criteria. Anyway here list the number of observations in the years_father data that have negative values. 2 points

QUESTION 17 Now its time to reveal the main purpose of this investigation. A baby is deemed to have low birth weight (low_birth_weight) if it weighs less than 2500 grams (5.5 lbs) at birth. This is an important measure for the overall health of an infant: babies that are so small are at risk of a variety of problems (you can read more here if you are interested). Since the purpose of this research is to investigate factors linked to low birth weight we will go ahead and throw out any data that are not useful for this purpose. (When you are putting together your data for your own project you should trim the data so that everything in the paper reflects just these observations.) The command to toss some data is drop: if you use drop by itself then it will drop the variables listed after the drop command but if you do drop followed by if then it knows to get rid of observations meeting those criteria. So you can go ahead and drop if low_birth_weight<0 Stata should tell you that 139 observations were deleted by this command. Now lets see how things have changed. Above you were asked to investigate the number of years the mother has lived in her current neighborhood (years_home) given that she is married. What is the mean now? Be precise to at least 2-3 digits past the decimal. 2 points

QUESTION 18 One contributing factor to LBW (low_birth_weight) is when the mother is young. Lets make a new variable called young_mom that is 1 if mom_age is less than 19 (i.e. 18 or younger) and 0 otherwise. Then use tab with row to compare rates of LBW by mothers age like this: tab young_mom low_birth_weight row Enter here the % of young moms who had LBW babies. Enter the percent with at least two places after the decimal so if the right answer is 12.34% enter 12.34. 2 points

QUESTION 19 Oh shoot I forgot something. I hope you are using a do-file! If so this will be easy. Please go back and find out the median income (mom_inc) of the 139 observations missing information on low_birth_weight i.e. for low_birth_weight<0. Since we dropped them you cant find it out using the data that is currently loaded into Stata: with a do-file its a quick fix.

Requirements: 1 | .doc file