The Student Room Group

UCL Msc Economics 2012....Who is going?

Scroll to see replies

Reply 100
Original post by Econla
It looks like you are trying to convince yourself that you took the best option by choosing Cam instead of UCL.

Saying that UCL students would perform even poorly in Cambridge is not a polite way to say that Cambridge is maybe even rigurous than the first.

Maybe it is, but we could argue for years about who is the tough guy in the street and I don't understand what would be the sense for doing this.
....


just another one of those people who tries to use the Internet to "prove" they are superior in some way! Don't let it worry you, people who do this usually have some insecurity that comes out this way. You haven't been on TSR long, there are a couple of others around too :rolleyes:
Original post by janjanmmm
Here is a link to the similar paper from University of Edinburgh MSc Economics:

http://www.personal.dundee.ac.uk/~yfchen/SGPEmath/MathsClassExam_1011wCover.pdf

This is their directory, which, unlike UCL, is much easier to access:
http://www.sgpe.ac.uk/courses0910/


thanks very much, that is quite a resource there....a 140 page coursenotes pack online, a lot better than what I got at undergrad (and could probably expect as a masters student :P)
(edited 12 years ago)
Reply 102
I think you totally misunderstood what I am saying. I am talking about fact here and I am not talking about which programme is tougher or better than the others. Let me summarize the facts I am stressing.

1) Cambridge MPhil Economic Research covers more advanced stuff than UCL MSc Economics.

2) UCL MSc Economics cover more advanced stuff than the Cambridge MPhil Economics.

3) UCL MSc Economics students have very heterogeneous academic background. If you look at the comments of the marker: "Many also fail in trying to simplify a problem before actually solving it, thus finding themselves in complicated computations. Conditional expectations/distributions and the law of iterated expectations were not well understood", "Better understanding of how to address simple problems demanding the study of functions implicitly defined or how functions vary with variables or parameters. A more consistent understanding of basic statistics concepts".

4) Students at the UCL MSc Economics have very heterogeneous academic background as the MSc Economics is not as selective as Ox, Cam and LSE.

5) From my personal experience, people who do not have the right background could well be admitted by the UCL MSc Economics and such people are around 30-40% of the entire course.

This leads to the conclusion that the failure rate should be higher if they do Cambridge Economic Research as there were already 30% of the people failed the entire degree at UCL last year.

This also makes me to suggest that before coming to UCL people need to have a careful consideration of whether they can actually handle the course.

Econla. I am not so sure about your background and other people's background. But if juding by the 30% failure rate, people need to consider whether they will eventually end up in the 30% of failing group and make decision on whether it is worth to invest the money and time for the MSc Economics.
(edited 12 years ago)
Reply 103
So are you suggesting that what the markers saying "basic, simple" are actually not basic and simple O_o

And the people who got 98% in the exam are some mathematical wizards while the people who got 18-30 are those ordinary intelligent people who could not cope with "basic" and "simple" maths and stats O_o
(edited 12 years ago)
Reply 104
What do you mean UCL is not selective ah ah :angry:....

WTF Lalafell!!!!!!!...........UCL is not selective??????...........Let me tell you that I come from a university ranked 879 in the world, I studied Ornithology but I took Algebra (Sets and even I saw Quadratic equations) and Statistics for animal sets plus one course in economics (Economic History) and my final mark was 40/100, ranked 76 among 115 students. I am proud of my achievements!!!!

Are you saying me that UCL is not selective????......I was admitted, what else do you want to proof the opposite!!!!:mad:






jijiijij.....just kidding :tongue:

No, I just would like to see the Cambridge outcome on the same kind of exam.

If you are just claiming that UCL is not as selective as Cambridge....I can give you that. In fact Cam requires GRE, UCL doesn't. Starting just by this point, you can say that Cambridge is more selective.

Please do not continue comparing apples with oranges (Mphil RESEARCH with a "vanilla" Msc in Economics) despite for what I know, the only difference between MPhil Research and MPhil is just the dissertation.

Anyway, if you are saying that Cambridge is more selective...oh yes I'm pretty sure they are.

Everybody knows that Harvard is more selective than Chicago (I've seen many people admitted to Chicago and rejected in Harvard). But can you say that the programme in Harvard is more rigorus than the last??....I think we can't. Both are top schools and their output is (quality of graduates) beyond an average school, so if there is a difference between choosing H or C, I think is marginal and subject just to brand, where Harvard as in the case of Cambridge with UCL wins.

Finally, I can not infer your 30% here and there etc......to do that we would need to compare both exams. There you can infer something. Not neccessarily for being more selective, a programme is harder or cover more advanced stuff.

The MPhil research has more compulsory courses than MPhil, hence is more strict, but without comparing real exams you can not imply that x number of UCL students would fail the other course.

But I give you the issue that Cam and Ox are more selective perhaps than LSE, UCL or Warwick.
Reply 105
Original post by Lalafell
I think you totally misunderstood what I am saying. I am talking about fact here and I am not talking about which programme is tougher or better than the others. Let me summarize the facts I am stressing.

1) Cambridge MPhil Economic Research covers more advanced stuff than UCL MSc Economics.

2) UCL MSc Economics cover more advanced stuff than the Cambridge MPhil Economics.

3) UCL MSc Economics students have very heterogeneous academic background. If you look at the comments of the marker: "Many also fail in trying to simplify a problem before actually solving it, thus finding themselves in complicated computations. Conditional expectations/distributions and the law of iterated expectations were not well understood", "Better understanding of how to address simple problems demanding the study of functions implicitly defined or how functions vary with variables or parameters. A more consistent understanding of basic statistics concepts".

4) Students at the UCL MSc Economics have very heterogeneous academic background as the MSc Economics is not as selective as Ox, Cam and LSE.

5) From my personal experience, people who do not have the right background could well be admitted by the UCL MSc Economics and such people are around 30-40% of the entire course.

This leads to the conclusion that the failure rate should be higher if they do Cambridge Economic Research as there were already 30% of the people failed the entire degree at UCL last year.

This also makes me to suggest that before coming to UCL people need to have a careful consideration of whether they can actually handle the course.

Econla. I am not so sure about your background and other people's background. But if juding by the 30% failure rate, people need to consider whether they will eventually end up in the 30% of failing group and make decision on whether it is worth to invest the money and time for the MSc Economics.



Are you smoothing your opinions??? :rolleyes:......trying to be my friend now???:redface:....hahahha........relax if you are admitted to Cambridge I will be the first to clap you.

Regards
Reply 106
Well I have nothing to say. I can only wish you won't be one of the 30% crying in prof. Wendy carlin's office in July 2013...

Amazing, isn't it, statistically 1 in 3 of your future colleagues will have the fate in July 2013 :smile:
(edited 12 years ago)
Reply 107
Original post by Lalafell
Well I have nothing to say. I can only wish you won't be one of the 30% crying in prof. Wendy carlin's office in July 2013...

Amazing, isn't it, statistically 1 in 3 of your future colleagues will have the fate in July 2013 :smile:


Well thank you...I think I could manage :wink:
Reply 108
Original post by Lalafell
Well I have nothing to say. I can only wish you won't be one of the 30% crying in prof. Wendy carlin's office in July 2013...

Amazing, isn't it, statistically 1 in 3 of your future colleagues will have the fate in July 2013 :smile:


I'm surprised you haven't learnt, in your elementary statistics classes, that you cannot use just one sample year to estimate the probability of failure in another year :s-smilie:.
(edited 12 years ago)
Reply 109
Original post by sj27
I'm surprised you haven't learnt, in your elementary statistics classes, that you cannot use just one sample year to estimate the probability of failure in another year :s-smilie:.


Are you joking? I am so surprised a person who claimed his/she was working as an economist and who claimed she/ he could not attend cam because of funding don't know the basic concept of information set...

1) as an economist as you claimed yourself, how do economists forecast future event?..

2) what are the elements in an information set?

3) what is a information set?

4) what is an AR process?


Can you answer this question??.:s-smilie::s-smilie::s-smilie:

:o:o:redface:
Reply 110
Well, seems you don't understand let me post a link about basic concept of forecasting

http://pages.stern.nyu.edu/~churvich/Forecasting/Handouts/Chapt1.pdf

And let me write a basic AR moel for forecasting...assuming that failure rate follows a random walk process within a short term.

Yt = Yt-1 + ut

I hope you understand what Yt-1 here means. Otherwise drop me a message I will explain to you when I have time...........
(edited 12 years ago)
Reply 111
Original post by Lalafell
Well, seems you don't understand let me post a link about basic concept of forecasting

http://pages.stern.nyu.edu/~churvich/Forecasting/Handouts/Chapt1.pdf

And let me write a basic AR moel for forecasting...

Yt = Yt-1 + ut

I hope you understand what Yt-1 here means. Otherwise drop me a message I will explain to you when I have time...........


Oh please, grow up for god's sake. Do you think that an AR model is appropriate in every single circumstance? Do you understand when an AR model is appropriate to use? Can you give a good economic reason as to why it would be an appropriate one to use in this case? No, thought not.

If I were to say to you that the preceding 10 years had shown failure rates of say 5%, ranging between 1% and 10%, would you still say that the 30% of last year means that "statistically speaking" you can expect a 30% failure rate this year?

I strongly suggest that you do not do anything empirical in your dissertation at least until you have learned what type of model is appropriate for a given set of data.
Reply 112
Oh, my god... I so shock of such amerished statements

As a student at UCL for 3 years, I have good knowledge of the failure rate each year and they remain around 25-30% each year. Last year, there were only 10% of the people failed Maths and Stats test.

Economic reason to use AR model? are you joking again? As a rule of thumb. Simple ARMA model for forecasting short-run economic time series outperform any structural economic model. And tons of studies have shown that simple ARMA model outperform structural economc model in forecastign exhcnage rate movement, stock prices movement, interest rate, GDP and so on and so far.

How do people forecast US GDP next year? An ARMA process beats almost every structural economic models.
Reply 113
Do you need reference for what I am saying? but I think people should know these basic facts... particularly for an economist :frown:
Reply 114
Original post by sj27
Oh please, grow up for god's sake. Do you think that an AR model is appropriate in every single circumstance? Do you understand when an AR model is appropriate to use? Can you give a good economic reason as to why it would be an appropriate one to use in this case? No, thought not.

If I were to say to you that the preceding 10 years had shown failure rates of say 5%, ranging between 1% and 10%, would you still say that the 30% of last year means that "statistically speaking" you can expect a 30% failure rate this year?

I strongly suggest that you do not do anything empirical in your dissertation at least until you have learned what type of model is appropriate for a given set of data.


Wait? you are saying you test for the staitistical difference? how do you test the statistical difference of future event?

You don't know the future and you forecast the future and why do you test something you don't have actual data??

Didn't you elementary time series lectures teach you that the significance of coefficients or parameters are immaterial in forecasting?
(edited 12 years ago)
Reply 115
You provided one year of data... if you had provided more than that (as you suddenly now say you have) that would have been perfectly acceptable. See, you yourself are even now justifying it by providing more data rather than trying to say you can just use an AR process!!

Oh, and here's a hint: Cambridge might just teach you that you need theoretical reasons to justify things. You can't just say "ARMA forecasts US GDP better therefore I can use it to predict the failure rate at UCL" ....that is not an economic reason!!! An economic reason justifies why you use a particular model for a particular set of data in terms of the underlying theory. But if you want to spend the rest of your life using an ARMA model on every data set you come across, go ahead....

Oh, and don't think I give two hoots about your attempts to belittle me. You are absolutely clueless as to who I am, who I work for, who I know, and who I interact with; I have plenty of reason to be very confident about my ability as an economist .... You can go ahead and make yourself sound petty if you want, but in a few years' time you'll back on this conversation in shame. Guaranteed. But I'm not going to bother continuing it.
Reply 116
...

Oh I am shocked again. I gave you a AR model and I suggested an information set. I assume that you know what I meant by AR model and information set. As the information set for forecasting contains all past and present values and AR model cannot be fitted using only one year data. But actually, I assume a random walk in the short run, this means that the time series for failure rate is a martingale so expectation of failure rate in year t conditional on past information set is the failure rate in time t.

Why this is not an economic reason? this is an emprical fact that maintained by emprical evidence. As such, using ARMA model for short run forecasting is economically and emprically justified due to the serial correlation of time series in different periods.

You might ask how do I assume for such serial correlation. Well, well well... the admission process, the same admission tutors over the past few years, the general admission rules all give serial correlation of the failure rate at different time periods within shorthorizon...

Do you have anything to say?
Reply 117
In any case. I think people misunderstood my intention of comparing the students at Cam and UCL and stressing the difficulties and failure rate of doing the UCL MSc Economics.

My past expereince told me that those people who are not qualified but got admitted are suffering throughout their studies at UCL. Having spending tremendous amount of time and money in the degree, those people at the end end up with nothing and had to resit in the subsequent year. Some even failed again in the resit and end up with no degree havng spending such time and money.

My intention is that, let me make it directly, don't come if you don't have the background... it is a waste of time and money and those people will suffer for sure throughout the entire year doing something not suitable.
Reply 118
Original post by sj27
You provided one year of data... if you had provided more than that (as you suddenly now say you have) that would have been perfectly acceptable. See, you yourself are even now justifying it by providing more data rather than trying to say you can just use an AR process!!

Oh, and here's a hint: Cambridge might just teach you that you need theoretical reasons to justify things. You can't just say "ARMA forecasts US GDP better therefore I can use it to predict the failure rate at UCL" ....that is not an economic reason!!! An economic reason justifies why you use a particular model for a particular set of data in terms of the underlying theory. But if you want to spend the rest of your life using an ARMA model on every data set you come across, go ahead....

Oh, and don't think I give two hoots about your attempts to belittle me. You are absolutely clueless as to who I am, who I work for, who I know, and who I interact with; I have plenty of reason to be very confident about my ability as an economist .... You can go ahead and make yourself sound petty if you want, but in a few years' time you'll back on this conversation in shame. Guaranteed. But I'm not going to bother continuing it.


Sorry, I missed the last bit. You said I am belittling you. But you were the first to start such provocative statements saying something elementary statistics and you kept attacking me directly or indirectly here or other posts. I remember in an post about Oxford MPhil, I am commeting on the course content with someone else but you came and started some provocative statements without understanding my posts...

I don't know who you are, I am sharing my expereince with other people. If you did not start such provocative statements consistently here and there, I would have not attacked you.
Original post by sj27
I'm surprised you haven't learnt, in your elementary statistics classes, that you cannot use just one sample year to estimate the probability of failure in another year :s-smilie:.


I would not expect the rates of failure to be very different from year to year, though. They are using the same method to select people, statistically the cohort would be very similar, why would failure rates range? Most admission statistics have upward or downward trends, not fluctuations. There is not enough evidence to suggest a trend, so if I had to guess based on one year I would say that it will, indeed, remain stable (but with low level of confidence, of course).
(edited 12 years ago)

Quick Reply

Latest

Trending

Trending