It is phone screening, and asked me two questions which are put below. The interviewer is nice, and kindly describe the problems. I suppose the first question is very tricky, I involved in Bayes formula for a long time, so there is not much time for the second one. After the interview, I realized that there is no need to use Bayes formula at all.
Interview questions [1]
Question 1
1. 100 students, make survey cheat/not cheat, student toss a coin, if head, tell truth, if tail, toss again, if Head then yes, if tail then no. 30 students answer yes, Q: how many student cheated?
2. Given an array, use bootstrap method to get the var of mean (I used R).
3. Can you write 2 in one line within function? ( so use apply() )
I applied online. The process took 3 weeks. I interviewed at Google (Mountain View, CA) in Mar 2016
Interview
I applied online and got an email from recruiter few days later. The phone interview was 45 minutes. The interviewer introduces his team and him self. He asked me statistical questions.
I had onsite interview at Mountain view. It took about 4 hours and I met 4 interviewers. They asked programing and statistical questions.
Interview questions [1]
Question 1
statistical modeling, programming, AB testing, probability
I applied online. The process took 2 weeks. I interviewed at Google in Mar 2016
Interview
I applied online and got an email from recruiter few days later. Scheduled a phone interview afterwards. The interviewer first introduces his team and then go on asking me statistical questions. I didn't know the answer and struggle for a while. Then he moves on to ask me to explain the algorithms I used before. The phone interview was 45 minutes.
Interview questions [1]
Question 1
Suppose N students participate in a coin flip experiment, when they get heads they stop, when they get tails they keep going. All students will stop after the second trial no matter the results. Y is the binary indicator of whether they claim they cheated in the experiment. Estimate how many students cheat in this experiment.