How do I reach the entry standard for a Masters in Statistics?

Watch
matthewneave
Badges: 2
Rep:
?
#1
Report Thread starter 1 year ago
#1
I'm pushing 40, graduated with a Natural Sciences degree in 2004, and have been out of academia for a good 15 years. A combination of work (consulting/finance) and a rediscovered love of numbers in middle age has meant that I'd like to do a masters in statistics at some point, but I think I'm some way off the start line (I'm using the Imperial College course as a guide).

With skill fade, my current level of stats is probably between A Level and fresher, while my calculus, vectors and matricies is better. Being a scientist I avoided/wasn't taught a lot of pure maths e.g. the rigorous mathematical proof process is quite unfamiliar to me.

My goal is to really understand the maths behind a lot of the regression analysis and other statistical techniques that I've encountered at work, and on the Python and R courses that I'm now learning. I'm quite prepared for this process to take years, so I'm not expecting a quick fix (although if one exists then great!).

Ultimately, my main aim is to understand the maths. The desire to pick up a masters is to have that recognised on my CV, but also because I think I would enjoy it. I'm interested in other branches of maths only in as much as it helps me on the stats front, hence not wanting to do a full blown undergrad maths degree first.

For example, one recommended read on the MSc pre-course reading list was 'Probability and Random Processes' by Gimmett and Stirzaker, but it feels like I'm missing a reasonable amount of prerequiste maths to understand it. Does this book typically come in part way through an undergrad course, and if so what ground do I need to cover first?

I've bought some books recommended on Imperial's undergrad reading list for A level students about to start, namely Elements of Logic via Sets and Numbers" by D L Johnson and "A First Course in Probability", Sheldon Ross. Is this a good start point to dust off cobwebs?

Broad question I know, but thoughts and opinions would be very welcome!
1
reply
alleycat393
Badges: 21
Rep:
?
#2
Report 1 year ago
#2
My first question is why? Unless you plan to move into a stats role, why do you want to understand the maths at masters level? If you plan to carry on as a scientist then understanding the applications of your stats is enough rather than needing the pure maths.

That aside if you want to do this then the short courses are a good place to start. You may also want to ask Imperial how they would see you as an applicant now and what you would need to meet their requirements. They may not be willing to consider you without the maths background.
(Original post by matthewneave)
I'm pushing 40, graduated with a Natural Sciences degree in 2004, and have been out of academia for a good 15 years. A combination of work (consulting/finance) and a rediscovered love of numbers in middle age has meant that I'd like to do a masters in statistics at some point, but I think I'm some way off the start line (I'm using the Imperial College course as a guide).

With skill fade, my current level of stats is probably between A Level and fresher, while my calculus, vectors and matricies is better. Being a scientist I avoided/wasn't taught a lot of pure maths e.g. the rigorous mathematical proof process is quite unfamiliar to me.

My goal is to really understand the maths behind a lot of the regression analysis and other statistical techniques that I've encountered at work, and on the Python and R courses that I'm now learning. I'm quite prepared for this process to take years, so I'm not expecting a quick fix (although if one exists then great!).

Ultimately, my main aim is to understand the maths. The desire to pick up a masters is to have that recognised on my CV, but also because I think I would enjoy it. I'm interested in other branches of maths only in as much as it helps me on the stats front, hence not wanting to do a full blown undergrad maths degree first.

For example, one recommended read on the MSc pre-course reading list was 'Probability and Random Processes' by Gimmett and Stirzaker, but it feels like I'm missing a reasonable amount of prerequiste maths to understand it. Does this book typically come in part way through an undergrad course, and if so what ground do I need to cover first?

I've bought some books recommended on Imperial's undergrad reading list for A level students about to start, namely Elements of Logic via Sets and Numbers" by D L Johnson and "A First Course in Probability", Sheldon Ross. Is this a good start point to dust off cobwebs?

Broad question I know, but thoughts and opinions would be very welcome!
0
reply
matthewneave
Badges: 2
Rep:
?
#3
Report Thread starter 1 year ago
#3
Thanks for the reply. The 'why' is in large part to protect my career prospects, but also because it genuinely interests me in a way that it didn't when I was younger. Part work, part hobby.

Stats is becoming much more important in my industry (business intelligence and consulting). You could do a lot with a pretty basic set of mathematical tools, but as massive datasets become more accessible you can see how a lot of jobs are going to change or become obsolete in the next decade. My old firm is starting to move towards R, and you can see a real niche for people with a good set of soft skills and commercial experience coupled with a solid background in the maths.

I've hit on an MSc because its a well known qualification that appears in a lot of job specs, but the short course idea sounds interesting as a start point. Did you have anything specific in mind? Thanks again!
(Original post by alleycat393)
My first question is why? Unless you plan to move into a stats role, why do you want to understand the maths at masters level? If you plan to carry on as a scientist then understanding the applications of your stats is enough rather than needing the pure maths.

That aside if you want to do this then the short courses are a good place to start. You may also want to ask Imperial how they would see you as an applicant now and what you would need to meet their requirements. They may not be willing to consider you without the maths background.
0
reply
alleycat393
Badges: 21
Rep:
?
#4
Report 1 year ago
#4
Right so there are two stories here:
1. How your job will change and evolve over the next 10 years which I think will be something to watch as employers will upskill their employees based on what they need. The point is still that with using things like R, which is a software package, you need to understand how to interpret and use your results, not necessarily understand the basic maths behind them.
2. Where you want to be with maybe a new job that needs a stats degree.

I think your first step is to decide which of the above two routes you want to take. If it's 2. then see what Imperial would like to see in an application and go from there.

Re: short courses, these may be offered on software packages like R. Have a look online.
(Original post by matthewneave)
Thanks for the reply. The 'why' is in large part to protect my career prospects, but also because it genuinely interests me in a way that it didn't when I was younger. Part work, part hobby.

Stats is becoming much more important in my industry (business intelligence and consulting). You could do a lot with a pretty basic set of mathematical tools, but as massive datasets become more accessible you can see how a lot of jobs are going to change or become obsolete in the next decade. My old firm is starting to move towards R, and you can see a real niche for people with a good set of soft skills and commercial experience coupled with a solid background in the maths.

I've hit on an MSc because its a well known qualification that appears in a lot of job specs, but the short course idea sounds interesting as a start point. Did you have anything specific in mind? Thanks again!
0
reply
threeportdrift
Badges: 20
Rep:
?
#5
Report 1 year ago
#5
(Original post by matthewneave)
I'm pushing 40, graduated with a Natural Sciences degree in 2004, and have been out of academia for a good 15 years. A combination of work (consulting/finance) and a rediscovered love of numbers in middle age has meant that I'd like to do a masters in statistics at some point, but I think I'm some way off the start line (I'm using the Imperial College course as a guide).

With skill fade, my current level of stats is probably between A Level and fresher, while my calculus, vectors and matricies is better. Being a scientist I avoided/wasn't taught a lot of pure maths e.g. the rigorous mathematical proof process is quite unfamiliar to me.

My goal is to really understand the maths behind a lot of the regression analysis and other statistical techniques that I've encountered at work, and on the Python and R courses that I'm now learning. I'm quite prepared for this process to take years, so I'm not expecting a quick fix (although if one exists then great!).

Ultimately, my main aim is to understand the maths. The desire to pick up a masters is to have that recognised on my CV, but also because I think I would enjoy it. I'm interested in other branches of maths only in as much as it helps me on the stats front, hence not wanting to do a full blown undergrad maths degree first.

For example, one recommended read on the MSc pre-course reading list was 'Probability and Random Processes' by Gimmett and Stirzaker, but it feels like I'm missing a reasonable amount of prerequiste maths to understand it. Does this book typically come in part way through an undergrad course, and if so what ground do I need to cover first?

I've bought some books recommended on Imperial's undergrad reading list for A level students about to start, namely Elements of Logic via Sets and Numbers" by D L Johnson and "A First Course in Probability", Sheldon Ross. Is this a good start point to dust off cobwebs?

Broad question I know, but thoughts and opinions would be very welcome!
Have you looked to see what courses the OU offer? The often have a range of 'work up' courses that might suit, a year doing an introductory course, a year doing an undergrad module and then look at brick Uni's for a full Masters? The OU would also resolve most Uni's concerns about being out of education for so log - they often want evidence of undergrad level study within the last 3 years.
0
reply
freerange
Badges: 8
Rep:
?
#6
Report 1 year ago
#6
I'd also take a look at the courses and programmes available through edx.org, coursera.org & possibly futurelearn.com . You can audit most of them for free, or pay if you want a certificate for completing them.
0
reply
matthewneave
Badges: 2
Rep:
?
#7
Report Thread starter 1 year ago
#7
Thanks for advice everyone, really helpful!
0
reply
X

Quick Reply

Attached files
Write a reply...
Reply
new posts
Back
to top
Latest
My Feed

See more of what you like on
The Student Room

You can personalise what you see on TSR. Tell us a little about yourself to get started.

Personalise

If you don't put your camera on in online lessons, why is that?

My teacher doesn't want us to (75)
16.67%
No one else does (141)
31.33%
I'm embarrassed about my background (50)
11.11%
I feel self-conscious showing my face (155)
34.44%
We don't use a video platform (7)
1.56%
I don't have a camera (10)
2.22%
Something else (tell us in the thread) (12)
2.67%

Watched Threads

View All