## Difference between applied statistics and mathematical statistics.

What is the difference between applied and mathematical statistics ?

Does applied statistics skip a great deal of necessary knowledge ?

Is applied statistics enough to construct statistical models?
 PhysOrg.com science news on PhysOrg.com >> New language discovery reveals linguistic insights>> US official: Solar plane to help ground energy use (Update)>> Four microphones, computer algorithm enough to produce 3-D model of simple, convex room
 Are you referring to university classes? They probably differ between schools, so I'm not sure you'll get an accurate answer. In the workplace I've never heard anyone distinguish between those two things - you're either doing statistical modeling or you're not. The reality is you won't learn how to do it well in university classes, period, so just keep that in mind.
 I have a job but the analysis is mostly just data extraction and the models are just simulation (or classical models). I would not call it statistical analysis or modelling but I don't really know what I am talking about either. We've got people like me without much of an education and then we've got people with masters degrees in engineering and even though I consider that my work place is full of incompetents, the people with the strong mathematical background produce better solutions to problems. To be blunt, I can't see how another mickey mouse degree (only this time in 'statistics') is goign to help. I don't really know what 'mickey mouse' would be either. If I need to know how to do something I just look it up and there is usually a computer package that does it for me. http://en.wikipedia.org/wiki/Exponential_smoothing http://en.wikipedia.org/wiki/T_test I like this sort of basic data/statistics stuff, so I'd like to keep doing it but I don't really know anything about it beyond my first job in the field. I don't think I'd be highly employable without at least some sort of math or stat degree and it is bad to have all your eggs in one basket by depending on having the one job forever.

## Difference between applied statistics and mathematical statistics.

I work as data analyst in a business setting and I can tell you that while useful, statistics training is just too far out there for ordinary projects. Lots of time you need to perform data extraction and have solid knowledge of what you're extracting from the system. This is valued far more than ability to use package (like say R or SAS) and analyze for distribution of some metric, say. This kind of approach is suitable for a Ph.D. level researcher, and then you're just talking about a different caliber job all together.

 Quote by Monte_Carlo I work as data analyst in a business setting and I can tell you that while useful, statistics training is just too far out there for ordinary projects. Lots of time you need to perform data extraction and have solid knowledge of what you're extracting from the system. This is valued far more than ability to use package (like say R or SAS) and analyze for distribution of some metric, say. This kind of approach is suitable for a Ph.D. level researcher, and then you're just talking about a different caliber job all together.
Thanks for the advice. In terms of data extraction and manipulation SAS is very similar to SQL. Infact you can write SQL code in SAS using proc SQL, which is the preferred way of extracting DATA. It is also very easy to output data into a presentable format using SAS ODS.

I do a lot of extraction and manipulation of data.

What you are describing sounds like the kind of career path I want to follow. I fell into this job and I like it and it pays well so I want to be able to move onto new jobs and stay employable. For this I think I need some quantitative tertiary qualification because most adverts will specify that as a requirement.

How do you extract data ?

For roles such a Credit risk analyst or portfolio analyst is this mostly heavy math, or data analytics based ?
 Well to be honest I'm just trying to find my way in the big bad world. I've always liked numbers and math but I get confused with complicated concepts and I've never taken mathematics at university level because I've never had the right amount of time to dedicate to it. Now that I've got a more stable well paying fulltime job I've been thinking about taking maybe one or two units a term. My job is mostly data extraction and manipulation using SAS/SQL. From this data we develop reports for various people and data extractions based on requests from within the organisation. We use the data to determine future costs and revenues. This involves using basic forecasting techniques, and extraction and manipulation to get to more of a best guess, on my side of things anyway. A lot of it really just is pull the numbers out of the data-base, make it into a pretty table in excel, write some words, do some more tables and email it. I spend a lot of my time either wondering how the hell I'm going to get my work done in the small time frame or being bored and learning new and interesting ways to do things better. I like watching how the data collapses into usable information and seeing the trends and knowing what the results will be in the end so I want to stick to this sort of work. I've seen there are various jobs in finance such as credit risk modelling or portfolio analysis. I'd like to get into something like that in the next 3-5 years if I've got the ability. Which is where I am stuck on if I should get a tertiary and what I should do and if I have the ability to do it. How I ended up in this role was pure chance. I got a job doing some other **** this one opened up and I applied for it. The people doing the hiring knew me, they liked me and knew I was the kinda person who would enjoy this sort of work so they hired me. But yes, I am really just trying to nut out some basic skills and stay employed.
 I think your future in this line of work depends on reasons you signed up for it. In my case, working with numbers in Excel/VBA and massaging data in Access/SQL Server was far more to my liking than being on the phones in Cust. Serv. Also, it looks more natural with my biology background, although many ordinary people don't make a distinction between analytical and life sciences. When I find myself having to conjure up a nice story explaining how I wound up in my role, I try to invoke an idea that it's appropriate for modern biologist to be familiar with the techniques of data analysis. Of course I know I'm prevaricating - biology has little to do with data analysis - but at least it puts my academic background into some perspective. Down the road I see dim prospects. Most people in management are there for management, i.e. they see it as an ultimate fulfillment of their career aspirations. Certainly I don't see management in that way. I find my work rather circumstantial to the enterprise of backstabbing, gossiping and compelling peers to see numbers and charts in a way advantageous to some big shot. I'd much rather work with financial transactions because there is a meaning to them: finance is the lifeblood of any business. In my opinion, SAS training confers an advantage, but you've got to stay mobile: many jobs in science/engineering/technology require relocation to where the work is. I don't know if you're in Australia, but here in the states, it's pretty much a reality. Since you're around 27, you should be ok with that, but as you get older, it can become a problem. I'd suggest staying away from long-term financial obligations, e.g. debts, mortgages, etc that are unrelated to education. Basically you need to work hard and get lucky at the same time. Sounds like you're interested more in the scientific/technical part of it all (like myself.) If that's the case, you'll soon realize that without Ph.D. and experience there is a glass ceiling. Most quant jobs I see require Ph.D. in Comp. Sci. / Math / Physics and years of experience. Perhaps you should consider career in clinical data analysis for pharmaceutical trials. Let me know your thoughts on this.

 Quote by Monte_Carlo I think your future in this line of work depends on reasons you signed up for it. In my case, working with numbers in Excel/VBA and massaging data in Access/SQL Server was far more to my liking than being on the phones in Cust. Serv. Also, it looks more natural with my biology background, although many ordinary people don't make a distinction between analytical and life sciences. When I find myself having to conjure up a nice story explaining how I wound up in my role, I try to invoke an idea that it's appropriate for modern biologist to be familiar with the techniques of data analysis. Of course I know I'm prevaricating - biology has little to do with data analysis - but at least it puts my academic background into some perspective. Down the road I see dim prospects. Most people in management are there for management, i.e. they see it as an ultimate fulfillment of their career aspirations. Certainly I don't see management in that way. I find my work rather circumstantial to the enterprise of backstabbing, gossiping and compelling peers to see numbers and charts in a way advantageous to some big shot. I'd much rather work with financial transactions because there is a meaning to them: finance is the lifeblood of any business. In my opinion, SAS training confers an advantage, but you've got to stay mobile: many jobs in science/engineering/technology require relocation to where the work is. I don't know if you're in Australia, but here in the states, it's pretty much a reality. Since you're around 27, you should be ok with that, but as you get older, it can become a problem. I'd suggest staying away from long-term financial obligations, e.g. debts, mortgages, etc that are unrelated to education. Basically you need to work hard and get lucky at the same time. Sounds like you're interested more in the scientific/technical part of it all (like myself.) If that's the case, you'll soon realize that without Ph.D. and experience there is a glass ceiling. Most quant jobs I see require Ph.D. in Comp. Sci. / Math / Physics and years of experience. Perhaps you should consider career in clinical data analysis for pharmaceutical trials. Let me know your thoughts on this.
Sounds like you paint a pretty dim picture. The job market might be somewhat different in Australia as I understand the US has around 10% unemployment. What you are describing makes me seem content to stay in the my job for a long time as it is secure and interesting.

But as I said, I fell into this role, it is a job which I rather enjoy but I haven't / am cautious to invest a large amount of eduction into it.

Not sure what I'll end up doing long term, hopefully this, maybe something else.

Working in an office is good. You get used to the easy work and high pay which might make it harder to transition to other jobs.
 Working in an office environment confers a false sense of security. You should visit bowels of the organization and remind yourself that one day you may end up there. After reviewing SAS job market, I'm under the impression that, at least in US, most employers want an experienced SAS user. However, SAS training is far from free. I've heard that some companies hire data analysts and make them learn SAS in-house. Of course the costs of training are incurred by the company, not employee. What was your SAS experience prior to filling the current role? Did you learn SAS on-site or did you obtain the training someplace else?

 Quote by Monte_Carlo Working in an office environment confers a false sense of security. You should visit bowels of the organization and remind yourself that one day you may end up there. After reviewing SAS job market, I'm under the impression that, at least in US, most employers want an experienced SAS user. However, SAS training is far from free. I've heard that some companies hire data analysts and make them learn SAS in-house. Of course the costs of training are incurred by the company, not employee. What was your SAS experience prior to filling the current role? Did you learn SAS on-site or did you obtain the training someplace else?
SAS is not a terribly difficult language to learn. You can easily learn it within about a month but still not master it after 10 years.

 Similar discussions for: Difference between applied statistics and mathematical statistics. Thread Forum Replies Academic Guidance 4 Career Guidance 0 Academic Guidance 0 Academic Guidance 2 Academic Guidance 2