Tips On Getting Started In Python

whether that is good or not is of course mostly a function of the quality of the IT team. A fool-proof front-facing app is in my opinion much more stable and better than letting traders work with Excel Add-ins which are often unstable. I used an app in one of the previous banks, I traded at, which submitted requests to an in-house server and returned analytics, computed utilizing market data and other sentiment data. It was a fantastic app and worked all the time unless their internal market-data server had issues. The few times there were outages it was generally an issue on the market data vendor side (RMDS and Bloomberg) and not on the bank's end. On the other hand, I frequently had issues as trader with the derivatives pricer add-ins that sometimes "magically" disappeared from the Excel app or had issues connecting to messages buses or the likes. As soon as an RMD server is involved or UDF functions in Excel add-ins you are asking for trouble. At least that has been my experience...


sle said:
Yup, they do. In most cases it would be better if they did not :)
More...
 
For charting i would use low level or higher level lib what also allows low level custom drawing.
Then it becomes possible to visually render algos work, see behind the scenes and visually debug for problems or improvements.
 
Zzzz1 said:
The lecturer forgot to mention another reason why the CIS dept at Stanford has more honor code and plagiarism violations than all other departments combined is to a large degree Asian students. ........
Sorry, unrelated to Python but as the posted lecture did not focus on Python but honor code I thought I wanted to inject this point. It equally pertains to trading and finance in general. There is so much blatant cheating going on in stock markets around the world and nobody really is competent enough out there to properly police it.
More...

Blah, blah, blah. Do you expect us to believe such cheating doesn't occur on other US or European campuses? I will grant you there tends to be more cheating of the sort from certain ethnic backgrounds, you however paint a black or white moral situation when it relates to race.

You have a problem with morality and ethics in the world and finance? Go join the church.
 
Python charting libs are pretty shitty, but hey one can always use d3.js or something similar to run visualizations. No need to stare at matplotlib graphs.
 
Clearly you have not carefully read my post. I did not paint a black or white picture. I believe plagiarism is an issue that people need to be educated about and it has to be frowned upon in society in order for people to care about this issue. In Asia plagiarism is virtually a non issue. Even in 2016 you can find tons of counterfeit products on Taobao and Alibaba. Students copy each others homework and most universities in Asia simply do not care about it. Hence it does factor into the equation at US colleges and the higher the Asian student share of the class is the more issues there are relating to plagiarism and sharing of original work. I never even hinted at denying the fact that plagiarism exists and is an issue among all other ethnic groups. But facts clearly single out Asian students here. Some of the statistics that US university bodies, dared to publish, point to roughly 70%+ of all plagiarism cases to be conducted by Asian students. I could also second those numbers from 4 years in a US college and 2 more years in grad school. This problem is not just an issue at lower ranked colleges but goes all the way to the top, and probably more so at the top.

If you have something constructive to add then please do so other than mere name calling.

jj90 said:
Blah, blah, blah. Do you expect us to believe such cheating doesn't occur on other US or European campuses? I will grant you there tends to be more cheating of the sort from certain ethnic backgrounds, you however paint a black or white moral situation when it relates to race.

You have a problem with morality and ethics in the world and finance? Go join the church.
More...
 
Even d3.js is incredibly low performing and outright visually boring compared to some of the libraries existing in oop space. We are not talking some sweet looking pie charts with 5 different groupings. We are talking time series with potentially hundreds of thousands or millions of data points, the ability to swiftly pan and zoom, overlay, add or remove analytics, and all that at swift refresh rates. It simply is not the Forte of python and other interpreted languages and I don't even know of a simple java library that could easily handle hundreds of thousands of data points (which you get for a single day worth of tick data for the price series alone)

jj90 said:
Python charting libs are pretty shitty, but hey one can always use d3.js or something similar to run visualizations. No need to stare at matplotlib graphs.
More...
 
I found this site which is dedicated to introductory programming education, with an emphasis on python. Some of it is oriented towards kids and/or programming neophytes. There is also a heavy focus on scientific applications and data analysis in their more advanced courses.

www.nclab.com

I don't know if anybody has posted this or not but there is always Zed Shaw's classic introductory text:

Learn Python the Hard Way

https://learnpythonthehardway.org/

About LPTHW, lots of people absolutely love it and the author is almost a hacker stereotype. However it is very DIY. NClab has much more handholding but it's target demographic is mainly children.

I made a half ass attempt at learning python a couple of years ago but I'm making a New Year's resolution early this year, finish the intro courses at nclab.com.
 
Last edited:
tommo said:
Thanks I have come across this course and was considering it. I didn't know if it was worthwhile to learn Python fundamentals first via CodeAcademy or the course would take you from coding newbie and get you up to speed in a trading context
More...
tommo, I emailed the guy who runs this course and he recommended learning the basics on CodeAcademy first.
 
Last edited:
Zzzz1 said:
Even d3.js is incredibly low performing and outright visually boring compared to some of the libraries existing in oop space. We are not talking some sweet looking pie charts with 5 different groupings. We are talking time series with potentially hundreds of thousands or millions of data points, the ability to swiftly pan and zoom, overlay, add or remove analytics, and all that at swift refresh rates. It simply is not the Forte of python and other interpreted languages and I don't even know of a simple java library that could easily handle hundreds of thousands of data points (which you get for a single day worth of tick data for the price series alone)
More...
If the resolution is smaller than than the ammount of data points no need to render all seperately.

By zooming do you mean smooth zooming?
Like zooming an image on smartphone or stepped zooming?
I have not yet seen any other libs support that way other than my own libs.
 
You must log in or register to reply here.
Back
Top