Consider an urban area where 100 000 people travel to work; assume you possess the following...

 

Transcribed Image Text:

Consider an urban area where 100 000 people travel to work; assume you possess the following information about them: General information Mode Car Underground Bus Total Average number of cars per household 2.40 1.60 0.20 0.55 Population information Family income (1000$/year) Low (<25) Medium (25-75) High (>75) Total 0 63.6 6.4 0.0 70.0 Cars per household 1 15.9 3.7 2.4 22.0 Mean income (1000$/year) 120 60 10 25 2+ 0.0 2.4 5.6 8.0 Total 75.5 12.5 8.0 100.0 You are required to collect a sample of travellers to estimate a series of models (with a maximum of 8 parameters) which guarantee a negligible specification error if you have available at least 50 observations per parameter. You are also assured that if you take a 20% random sample of all travellers the error will be negligible and there will be no bias. You have also the following information: Hourly cost of an interviewer = $2 per hour Questionnaire processing cost = 0.3 per form Time required to classify an interviewee = 4 min Time required to complete an interview = 10 min You are also given the following table containing recommended choice-based sample sizes: Subpopulation size % to be interviewed <10,000 10,000 - 15,000 25 20 15 15,000-30,000 30,000 - 60,000 10 > 60,000 5 Choose the most convenient sampling method (either random, stratified or choice-based). Consider an urban area where 100 000 people travel to work; assume you possess the following information about them: General information Mode Car Underground Bus Total Average number of cars per household 2.40 1.60 0.20 0.55 Population information Family income (1000$/year) Low (<25) Medium (25-75) High (>75) Total 0 63.6 6.4 0.0 70.0 Cars per household 1 15.9 3.7 2.4 22.0 Mean income (1000$/year) 120 60 10 25 2+ 0.0 2.4 5.6 8.0 Total 75.5 12.5 8.0 100.0 You are required to collect a sample of travellers to estimate a series of models (with a maximum of 8 parameters) which guarantee a negligible specification error if you have available at least 50 observations per parameter. You are also assured that if you take a 20% random sample of all travellers the error will be negligible and there will be no bias. You have also the following information: Hourly cost of an interviewer = $2 per hour Questionnaire processing cost = 0.3 per form Time required to classify an interviewee = 4 min Time required to complete an interview = 10 min You are also given the following table containing recommended choice-based sample sizes: Subpopulation size % to be interviewed <10,000 10,000 - 15,000 25 20 15 15,000-30,000 30,000 - 60,000 10 > 60,000 5 Choose the most convenient sampling method (either random, stratified or choice-based).

Answer rating: 100% (QA)

ANSWER Based on the given information we need to estimate models with a maximum of 8 parameters and collect at least 50 observations per parameter to ... View the full answer