Statistics and physics?


by xdrgnh
Tags: physics, statistics
xdrgnh
xdrgnh is offline
#1
Nov12-12, 01:46 PM
P: 306
I'm currently taking a intro to statistics class. It's a quarter class so it just started a week ago. I got a lot going on this semester so after taking the class I feel I'll only have a computational aspect for statistics. I won't know much about the theory because I know I won't have much time to study it. Is only knowing a formulas and there meanings in statistics enough to get by a undergrad thermo and statistical physics class that I will take in the future.
Phys.Org News Partner Science news on Phys.org
Better thermal-imaging lens from waste sulfur
Hackathon team's GoogolPlex gives Siri extra powers
Bright points in Sun's atmosphere mark patterns deep in its interior
homeomorphic
homeomorphic is offline
#2
Nov12-12, 07:21 PM
P: 1,028
The term "statistical" physics is somewhat misleading. You need to know some probability, but it's not really statistical. The role of statistics in physics, as in other sciences, is predominantly for analyzing experimental data.
Jorriss
Jorriss is offline
#3
Nov12-12, 08:31 PM
P: 1,025
Quote Quote by xdrgnh View Post
I'm currently taking a intro to statistics class. It's a quarter class so it just started a week ago. I got a lot going on this semester so after taking the class I feel I'll only have a computational aspect for statistics. I won't know much about the theory because I know I won't have much time to study it. Is only knowing a formulas and there meanings in statistics enough to get by a undergrad thermo and statistical physics class that I will take in the future.
Quote Quote by homeomorphic View Post
The term "statistical" physics is somewhat misleading. You need to know some probability, but it's not really statistical. The role of statistics in physics, as in other sciences, is predominantly for analyzing experimental data.
Homeomorphic is right. It's predominantly probability theory.

There is probably much overlap between probability and statistics in an intro course. Are you covering combinatorics? Different distributions? What is the course covering?

Mute
Mute is offline
#4
Nov12-12, 10:44 PM
HW Helper
P: 1,391

Statistics and physics?


A basic understanding of probability theory is the minimum you need to know in order to do statistical physics, but if you can manage it I think taking a full statistics course is something every scientist should do. Analyzing data - simulation or experimental - is a vital skill, and the more of a command you have over data analysis the better the conclusions you can draw from your data. Again, this is true for both experimental and simulation data analysis.

Advanced statistical theory is not necessary for an undergraduate level, or perhaps even a basic graduate level, statistical physics course, but if you go on to study statistical physics in grad school I think it would be helpful.
xdrgnh
xdrgnh is offline
#5
Nov13-12, 12:27 AM
P: 306
Understand the basics of probability theory
 Summarize basic descriptive statistics of a given data set, both numerically and graphically
 Obtain a solid understanding of all standard discrete random variables, including Uniform, Bernoulli, Binomial, Geometric, Negative Binomial, Hyper-geometric, and Poisson random variables
 Obtain a solid understanding of all standard continuous random variables, including Uniform, exponential, Gamma, Chi-square, and Normal random variables
 Apply knowledge of random variables to solve real application problems

It's only a quarter long so that is why what it covers is limited. My problem is that I got a lot of stuff going on so I only got time to get a superficial understanding of the course material. Will that superficial plug and chug knowledge of statistics give me the tools to do well in statistical physics class. In physics class I never have a plug and chug mentality. It's only that I wanted to drop this course but can't because I class that I took was before which I thought was 4 credits is now 3.
Mute
Mute is offline
#6
Nov13-12, 02:31 PM
HW Helper
P: 1,391
Check the syllabus for your department's statistical physics/mechanics class. It will probably cover the basics of probability theory and statistics that you need for the course. If you have to drop your statistics course, you probably won't end up behind in stat mech, but if you don't have to drop it you might be able to stay a little bit ahead.
Jorriss
Jorriss is offline
#7
Nov13-12, 02:41 PM
P: 1,025
Quote Quote by xdrgnh View Post
It's only a quarter long so that is why what it covers is limited. My problem is that I got a lot of stuff going on so I only got time to get a superficial understanding of the course material. Will that superficial plug and chug knowledge of statistics give me the tools to do well in statistical physics class. In physics class I never have a plug and chug mentality. It's only that I wanted to drop this course but can't because I class that I took was before which I thought was 4 credits is now 3.
Keep in mind, more probability theory is better for statistical mechanics but most people who take statistical mechanics, be it at an undergraduate or graduate level, have not taken a dedicated course on probability or statistics. Math wise, this course is more than enough to survive a first course.


Register to reply

Related Discussions
Statistics and Physics Career Guidance 4
Physics Vs Statistics Academic Guidance 2
Statistics for physics student.. General Physics 1