Recent Posts

Pages: [1] 2 3 ... 10
1
Paper Replication and Project / Re: Question on citations
« Last post by bkane2 on April 27, 2017, 10:58:04 PM »
Thanks for the quick response!


Yeah, I'm using Turabian/Chicago style. So for example here's an excerpt from what I have so far (from the intro/summary portion):


"To construct a measure of obesity, the author uses the standard measure of body mass index (BMI), created from height and weight measurements. However, as height was not observed in the year 1990, DeBeaumont instead uses height information from 1982, which he justifies by reasoning that most women would have stopped growing by 18 years old, which is the minimum age a subject could have been in 1982."


Should I cite things such as this? I guess my concern is in the introduction where I'm summarizing his argument, I do this kind of thing a lot, so I'd probably end up with a ton of citations linking to the same paper.
2
Paper Replication and Project / Re: Question on citations
« Last post by Scott Onestak on April 27, 2017, 09:03:15 PM »
Hey,


I'm not positive what citation style you're using, but it sounds like Chicago to me.  No matter the citation style, you should always cite your source when you use a direct quote or are paraphrasing.  If just referencing, like "these results are consistent with the results found in this paper," I don't think a citation is necessary.  However, pulling any information or idea that is not your own should always be cited.


Here's a quote from the Chicago citation manual:


"...you should include a note (endnote or footnote) each time you use a source, whether through a direct quote or through a paraphrase or summary. Footnotes will be added at the end of the page on which the source is referenced, and endnotes will be compiled at the end of each chapter or at the end of the entire document."
[/size]
[/size]If you want to read more, this is a good resource: [/size]https://owl.english.purdue.edu/owl/resource/717/01/
3
Paper Replication and Project / Question on citations
« Last post by bkane2 on April 27, 2017, 08:54:31 PM »
Hi, I have a question regarding citing the paper that this project revolves around.


Currently I have a footnote citation only when I first introduce DeBeaumont's paper in the beginning of my paper (and will have it in the references page as well). However, I refer to and paraphrase from DeBeaumont's paper multiple times in my introduction where I summarize his arguments. I was wondering if I need to cite all of these instances, or if it suffices to cite his paper only the first time I introduce it?


Best,
Ben
4
Material / Re: Not Missing-at-random Data
« Last post by rkao on April 26, 2017, 07:42:26 PM »
Got it, thanks!
5
Material / Re: Not Missing-at-random Data
« Last post by Thomas VanDer Straaten on April 26, 2017, 05:02:10 PM »
Little typo: the missing indicator takes on a value of 0 if it is not missing.
6
Material / Re: Not Missing-at-random Data
« Last post by Thomas VanDer Straaten on April 26, 2017, 04:59:47 PM »
Hi,


Essentially, what we are doing here is we recognize data missing not at random is a problem and can't be ignored simply by deleting the observations for which the data is missing. For example, if we are conducting a survey and ask questions about mental health - some questions won't be answered, because of the respondent's mental health (such as depression ). But that's extremely relevant information, and simply removing the observation will harm the integrity of our sample.


So if you have missing information not at random for a covariate x2 (in a set of variables x1 to xk), instead of deleting the entire observation, we create a new variable xk+1 that takes on a value of 1 if x2 is missing, and x2 if it is not missing. For example, visually, see attached. We'd then impute missing values to a constant and include the missing indicator xk+1 in our regression.

7
Material / Re: A question about omitted variables
« Last post by Shenxiong on April 25, 2017, 08:22:04 PM »
Thank you very much!
make sense to me!
8
Material / Re: A question about omitted variables
« Last post by Alexis Orellana on April 25, 2017, 03:26:35 PM »
Hello,

In both cases your omitted variable is Experience. If you
only add the linear term "Experience" your specification assumes that the marginal effect (that is, the value of your coefficient) of 1 unit of experience on wages is constant.


If you include "Experience" and "Experience^2" (with coefficients b3 and b4), then the only difference with the previous case is that now you're claiming that the relation depends on the level of experience, because your marginal effect will be: b3+b4*Experience.

I hope this clarifies your question.

Best,
Alexis
9
Material / A question about omitted variables
« Last post by Shenxiong on April 25, 2017, 12:36:47 AM »
Hi~


  Suppose that the model is wage=b0+b1educ+b2drink+u
  and it is obvious that we omit the variable "Experience"
  However, since wage is not linear relation with "Experience" but "Experience^2", if we want to add experience into the model, we need add exper and exper2.
  So, in this case, the omitted variable is exper or exper2 or both?
 
  Thank you so much
10
Material / Not Missing-at-random Data
« Last post by rkao on April 24, 2017, 09:19:05 PM »
Hello,


Could you please clarify the bottom of page 5 in Notes 19?  If there is missing information for x2, then what does it mean to have xk+1i=1 if x2 is missing and xk+1i=0 if x2 exists?  Could you provide an example to illustrate this; how setting a dummy for x2 creates a proxy for finding the reason the information is missing?


Thanks,
Richard
Pages: [1] 2 3 ... 10