Hi,

I am wondering if someone could recommend software purely designed to teach mathematics as though they knew nothing.

I found software better than textbooks that tend to cram information, and my old professor that used to waffle on with a piece of chalk and a blackboard as though we were already mathematicians or as if she was alone.

I used Calmat. 16 years ago. It was perfect but does not run on modern OSes.

Seriously considering returning to college to do a masters, via distance learning but the subject that is maths is putting me off the monetary gamble.

This thread seems good but can't rely on the regs still being here in september. In fact I notice one is already missing.

Before worrying what is is. Simply treat one as an XOR and the other as a AND gate with an inverted input as though they are isolated. Then once you have you truth table, you can see what it is.

This got me through it
https://www.allaboutcircuits.com/textbook/digital/

Digital fundamentals by Thomas L. Floyd is second to none.

Depending on what ends you are looking for, the software out there ranges from okay to awful.

I am completely unfamiliar with Calmat, but a quick google search found a reference to it on page 3 of this document:

https://www.tandfonline.com/doi/pdf/...68776960040106

Apparently, it was a standard type of rote-practice/exercise type of program. You can get something similar to that using Khan Academy. It seems to be DOS-based, and if you have the software you can likely emulate it using DOSBox.

Masters in what? How core will the math skills be?

In engineering. I left EE and went into IT. This course is centred around embedded systems. My real passion. When my math lecturer was changed to a new one my grades declines and so did my interest. Until she gave me calmat. Ran on XP but that is the program in your link. I'll going from not thinking about maths in all these years and would going back to solving first and second order difference equations. Thinking about Z transform and Fourier analysis etc while working a job. Stuff I should know but don't really remember. I'm not so worried about the rest of the syllabus. I think I'll check out Khan academy and make a decision

Left EE to go into IT to work I mean. Graduated EE but scraped a pass in maths.Now that I'm out of EE so long if I were to lose my job I would be in a strange place. Like a jack of traits.

That's completely not what I was expecting given your initial description of what you were looking for. I thought you were looking for a basic algebra primer, like remembering how to solve for x in linear equations and things like that.

I'm not in engineering, but I would suspect that most of what you're looking for is straight computational proficiency as opposed to some sort of deeper understanding of how/why the transforms work the way they do. I would imagine that the internet has a ton of videos and example problems for Z transforms and Fourier analyses, and those resources were probably not nearly as abundant back whenever you were using XP.

I understand. Although I framed the question as such for good reasons which I don't want to get into here, becoming philosophical and hijacking the thread as much as I'd love to. Keep doing what you're doing ITT if it keeps you going because I'm certain you're a legend educator here to students.

Please, anybody can help me to know who has the edge between those 2 players in this game of dice?
Player1 has +1 on his roll. 2=3, 4=5 etc... but Player1 give 2/1 to Player 2.

My guess is Player 1 has still the edge. If so, what kind of odds Player 2 can request for fairer bet?


Thanks

What about ties?

Edit: The quick answer is that you just have to look at all the possibilities. One way is to put it into a chart. Put player 1's rolls across the top and player 2 down the side:

Then in each position you put the corresponding win or loss value. We can do this from P1's perspective. (+1 win, -2 lose, ? tie)

Then just add up all the values in the chart.

Google Sheets Question

I track my poker results in google sheets and was wondering if there was an easy way to create a typical poker graph. I've tried using the chart function but the graph I get takes each result as starting from zero instead of accumulating from the previous results. Does anyone know of a fix for this?

Thanks in advance and sorry if this is the wrong thread for this.
Sam

You want to make a new column that is the sum of all previous columns.


I went into cell B2 and wrote "=A2", in cell B3 I wrote "=A3+B2" and clicked cell B3 and dragged the formula down and it auto-populated the accumulation.

Awesome. Thanks Cueball

Not a homework question but maybe somebody can help me as my statistics classes were a long time ago:

I work in manufacturing, and I have an idea for a change that I want to proof. The scrap rate for the defect I am trying to improve was .5 percent. I do not know what the standard deviation is. How do I go about testing that it is indeed an improvement? Tests are time costuming and manual so I can't make the change, collect 1000s of samples and go from there as I usually do.



Sent from my SM-G955U using Tapatalk

How many tests can you do?



https://en.wikipedia.org/wiki/P-value

z test for proportions

http://sphweb.bumc.bu.edu/otlt/MPH-M...icalData2.html

https://onlinecourses.science.psu.ed...ts/proportions

https://en.wikipedia.org/wiki/Z-test



https://www.youtube.com/watch?v=Pkyb05DZdZg

more on what is null hypothesis testing etc

https://www.youtube.com/watch?v=VK-rnA3-41c

Some example of the way it works

https://www.youtube.com/watch?v=t09Vyd7H52A

https://www.youtube.com/watch?v=H-Nl-PpPOcU

https://www.youtube.com/watch?v=SHLSn3EHVuQ

Also keep in mind that suggesting improvement of a low already 0.5% rate is hard to do with few tests (ie small sample). You can only reject improvement hopes and indicate worsening with few tests not reject the null hypothesis that there was no improvement.

You don't have the raw data?

Without it, you are just going to have to check the new scrap rate and if it is lower than 0.5% with your new process, then you claim victory. If it isn't then you sweep it under the rug.

Do you have any idea what your current rate after improvements should be?

If your improvement is say to 0.4% then it will take forever to prove at a 95% level of confidence ie that the null hypothesis chance to be true is less than 5%. You are in need of thousands of tests.

If you are to 0.25% now it will be better but still will take over 1000-2000 samples probably.

If you relax confidence level to 90% it will get better but still depends on how much of an improvement you think you have.

The improvement can be very real and yet you would be unable to reject the null hypothesis so often without many many samples if its still close to 0.5%.

Try this for example.

say your improvement is now at 0.25%.

If you checked n samples the 95% confidence interval for null hypothesis at 0.5% is 0.005+-1.96*(0.005*.995/n)^(1/2)

So you need to have 1.96*(0.005*.995/n)^(1/2)<0.0025 at the very least. That leads to n>3057.

If you tried far less than that you would find even if at 50% reduction of scrap rate that you are unable to reject null hypothesis very often. I mean you would have the satisfaction the rate is lower on avg but not low enough to claim true progress at over 95% confidence. If 80% or 90% confidence is ok then probably 1000-2000 samples would do. There is no serious reason to have 95% if its not super important to respect the null hypothesis. But you would still have some risk of just being lucky in your rate improvement.

hi, hoping people can help..

i am looking to upgrade my math skills (first year univ and some 2nd year) and computer (1st year intro course) skills... note that in brackets are my background knowledge)..

anyway, i am looking to do combo of self-learn (books and MIT opencourse ware) and online/university courses.

my interest is time series an analysis and data analytics... not that interested in derivatives pricing but it may be useful and unavoidable.

i'm trying to figure out the different areas of university math....

so far, i see:

calculus 1/2/3/4 (single and multivariate)
probability
linear algebra
differentiation (not sure if it's eventually separate category from "calculus")
numerical computation (its own category??)

anything else i need on the math side?

the computer science side is much tougher/thornier. probably have to learn that at a university (at least initially).. might need to self-learn higher concepts as often courses are closed to non-computer majors.

thx in advance for any thoughts