point estimate of a population mean?

Discussion in 'General Discussion' started by ILoveDinobot, Apr 7, 2011.

  1. ILoveDinobot

    ILoveDinobot Arise Rodimus Prime

    Joined:
    Jan 19, 2006
    Posts:
    15,005
    News Credits:
    7
    Trophy Points:
    287
    Location:
    Long Island
    Likes:
    +36
    Twitter:
    Does anyone know what a point estimate of a population mean is and how to find it? I googled it and spent the last hour reading every link that came up in google. I still have no idea have to find this point estimate. I actually just pulled out a few strands of hair, tugging on my bangs. That's never happened to me before, so this is really driving me crazy!

    My stats professor is a guy from India and you can not understand a word he says, and he doesn't understand our questions at all. Someone asks a question and he answers with something that isn't even relevant to the question asked. Everyone in the class tries to make him understand the questions and he just like NEVER gets it. So I think the rest of the class is as lost as me.

    So anyone, PLEASE help me! ;_;
     
  2. Chaos Muffin

    Chaos Muffin Misadventure Veteran

    Joined:
    Mar 26, 2004
    Posts:
    28,751
    Trophy Points:
    322
    Likes:
    +4
  3. ILoveDinobot

    ILoveDinobot Arise Rodimus Prime

    Joined:
    Jan 19, 2006
    Posts:
    15,005
    News Credits:
    7
    Trophy Points:
    287
    Location:
    Long Island
    Likes:
    +36
    Twitter:
  4. tusko

    tusko Well-Known Member

    Joined:
    Dec 3, 2004
    Posts:
    4,727
    News Credits:
    12
    Trophy Points:
    257
    Likes:
    +185
    Twitter:
    Google+:
    Flickr:
    Don't take my answer as absolute truth.... but this terminology comes into play when making statistics of a subset of values.

    So if you have a process, and you measure a group of them.... from this group you get an estimate of the mean. This isn't the true mean of the whole population, but based on the sample you can figure out how close you are.

    So for example you want an average response from teens, you can't ask every single teenager your question to get the true mean response - so from your subset you get a point estimate. And you say its close to an average response based on how you determine the variation in the measurement.

    I would get confused about this all the time in stats class too.
     
  5. Chaos Muffin

    Chaos Muffin Misadventure Veteran

    Joined:
    Mar 26, 2004
    Posts:
    28,751
    Trophy Points:
    322
    Likes:
    +4
    i dunno
     
  6. llamatron

    llamatron Shut up, Nigel. TFW2005 Supporter

    Joined:
    Jul 11, 2002
    Posts:
    8,969
    News Credits:
    9
    Trophy Points:
    262
    Likes:
    +11
    WIKI SAYS: "In statistics, point estimation involves the use of sample data to calculate a single value (known as a statistic) which is to serve as a "best guess" for an unknown (fixed or random) population parameter."

    So I guess one example would be summation of values divided by number of samples?
     
  7. ILoveDinobot

    ILoveDinobot Arise Rodimus Prime

    Joined:
    Jan 19, 2006
    Posts:
    15,005
    News Credits:
    7
    Trophy Points:
    287
    Location:
    Long Island
    Likes:
    +36
    Twitter:
    Thanks, I'll see if I can get something from trying that.

    So it's just a guess on an average?
     
  8. Aernaroth

    Aernaroth <b><font color=blue>I voted for Super_Megatron and Veteran

    Joined:
    Mar 27, 2004
    Posts:
    23,507
    News Credits:
    2
    Trophy Points:
    347
    Likes:
    +136
    A point estimate is a single discrete value given as an estimate, rather than a range of estimated likely values. Thus, most estimates have a confidence interval, point or otherwise.
     
  9. ILoveDinobot

    ILoveDinobot Arise Rodimus Prime

    Joined:
    Jan 19, 2006
    Posts:
    15,005
    News Credits:
    7
    Trophy Points:
    287
    Location:
    Long Island
    Likes:
    +36
    Twitter:
    Yeah, I already found the confidence intervals, but I still can't figure out how to find the point estimate.
     
  10. tusko

    tusko Well-Known Member

    Joined:
    Dec 3, 2004
    Posts:
    4,727
    News Credits:
    12
    Trophy Points:
    257
    Likes:
    +185
    Twitter:
    Google+:
    Flickr:
    I think its as simple as an average with error bars on it.

    So if you have 25 different numbers. You can figure out the exact average/mean for those numbers.

    If you take any 5 of those 25 you can figure out the mean of the sample (estimate) with a confidence level (error bar) which should be comparable to the exact mean.
     
  11. tusko

    tusko Well-Known Member

    Joined:
    Dec 3, 2004
    Posts:
    4,727
    News Credits:
    12
    Trophy Points:
    257
    Likes:
    +185
    Twitter:
    Google+:
    Flickr:
    Its an average with a confidence interval (error bar) to tell you how fuzzy the number actually is.

    The issue I always had was that then you are asked to get the "true average" and I could never wrap my mind around how your ever going to know that number.
     
  12. tusko

    tusko Well-Known Member

    Joined:
    Dec 3, 2004
    Posts:
    4,727
    News Credits:
    12
    Trophy Points:
    257
    Likes:
    +185
    Twitter:
    Google+:
    Flickr:
    Oh by the way, I'm going back about 10 years here. Please correct me if I'm wrong!!!
     
  13. Aernaroth

    Aernaroth <b><font color=blue>I voted for Super_Megatron and Veteran

    Joined:
    Mar 27, 2004
    Posts:
    23,507
    News Credits:
    2
    Trophy Points:
    347
    Likes:
    +136
    Why don't you post the question, then we'll have a better idea what you're looking for.
     
  14. ILoveDinobot

    ILoveDinobot Arise Rodimus Prime

    Joined:
    Jan 19, 2006
    Posts:
    15,005
    News Credits:
    7
    Trophy Points:
    287
    Location:
    Long Island
    Likes:
    +36
    Twitter:
    1. Suppose you are interested in estimating the mean height of the population of people between the age of 12 and 14 years. Assume that the height follows a normal distribution with unknown population mean (μ) but with known standard deviation (σ) = 6 cm. A random sample of 31 people of the given age group is selected. The same mean height (x-bar) = 147.4 cm.
    1.1 What is the point estimate of μ?
    1.2 Construct a 95% confidence interval (CI) for μ.
    1.3 Construct a 99% confidence interval (CI) for μ.
    1.4 Construct the respective cures and identity the CI (95 & 99) areas as well as the rejection areas/critical regions.
    1.5 Interpret both the CIs.


    I can answer all the parts except 1.1 :( 
     
  15. Chaos Muffin

    Chaos Muffin Misadventure Veteran

    Joined:
    Mar 26, 2004
    Posts:
    28,751
    Trophy Points:
    322
    Likes:
    +4
    *head explodes*
     
  16. ILoveDinobot

    ILoveDinobot Arise Rodimus Prime

    Joined:
    Jan 19, 2006
    Posts:
    15,005
    News Credits:
    7
    Trophy Points:
    287
    Location:
    Long Island
    Likes:
    +36
    Twitter:
    :bay 
     
  17. Aernaroth

    Aernaroth <b><font color=blue>I voted for Super_Megatron and Veteran

    Joined:
    Mar 27, 2004
    Posts:
    23,507
    News Credits:
    2
    Trophy Points:
    347
    Likes:
    +136
    Yes, the point estimate they are looking for in this case is the mean, I think.
     
  18. lars573

    lars573 Well-Known Member

    Joined:
    Mar 29, 2007
    Posts:
    6,852
    Trophy Points:
    197
    Likes:
    +27
    Statistics is shit simple.
     
  19. ShortCircuit

    ShortCircuit Decepticon

    Joined:
    Nov 28, 2006
    Posts:
    1,698
    Trophy Points:
    126
    Likes:
    +0
    31 people in a classroom who are 2.5 inches apart in height would look like robots.
     

Share This Page