Stats Notes - Highlights for reference (Unofficial)

[Tejasvi Hegde]

Here is stats notes I had prepared for myself. Thought it may be helpful to many

This basically contains highlights and ready reference

 

For Maths notes, check 

https://groups.google.com/a/nptel.iitm.ac.in/g/ma1001-discuss/c/ubIcoEBawtg

ps: Let me know in case of any error  

[nikhil kambli]

thank you

[nikhil kambli]

do you have of computational ( i know it is supposed to be easy, but after the mock i need to get cracking)

[Tejasvi Hegde]

@Nikhil, For CT, all you need to do is to think 'very' logical and think the flow and effects.

If you get time, substitute some sample values to logic and try to evaluate the logic and see the result.

in CT, they are analyzing the capability of one's thinking than anything technical or bookish.

[Ranjit John]

THANKS A LOT Tejesvi.....Its a great work.

With regards,

Ranjit John

[Harikumar P]

Thank you very much

[CG]

Thanks for this Teasvi. This is a good reference pdf.

I had not made the connection between SSE and variance until I saw your notes.

[Tejasvi Hegde]

To be honest, even I didn't!

But now looks like everything is boiling down to sum of squared difference between value and mean!

[Mohan Palanisamy]

Thanks Tajasvi. Great job.

by the by, Is SSE is related the std deviation.

On Friday, November 13, 2020 at 3:21:21 AM UTC+5:30 Tejasvi Hegde wrote:

[Sherlock Holmes]

This is such a great work on your part.You have my sincere thanks for sharing it with us!!

[Tejasvi Hegde]

Updated version 1.1

I have updated with ready reference of formulas in one place and chart of relation

[Malabika Guha Mustafi]

Thank you very much.

[Dipak Patil]

Thank You So Much😊😊

[DHARMA TEJA GODUMALA]

Hi Tejasvi Hegde,

Your notes shows your whole hard work and interest towards the course. I appreciate your work,which will be useful to every student pursuing this course.

Just a small correction that I had noticed in your work is relating SSE and Variance.

They are not relatable because SSE deals with the observed and predicted values where as variance deals with the spread of data. 

Thank you for knowledge sharing.

On Friday, November 13, 2020 at 3:21:21 AM UTC+5:30 Tejasvi Hegde wrote:

[Tejasvi Hegde]

Thanks @Dharma Teja.

>> They are not relatable because SSE deals with the observed and predicted values where as variance deals with the spread of data 

This you are referring in the last section of relations of formulas ? 

I just created them for memorizing purpose that which formulas are used in other formula. It was not for theoretical relationship

Let me know if anything is giving wrong direction, (If you can specify page and sentence, that will be helpful) I will update them.

Yes I do have interest, but Maths and Stats (especially looking at Mock Test) in Qualifier exams will determine my journey in this course :)

[CG]

Hi Tejasvi,

Maybe you could include computational formulae for standard deviation (population and sample). This will make it easier and faster for those struggling with calculations.

Population variance.

For Sample Variance

I was not aware of this reduction/simplification of the formulae until it was pointed out by someone else on this forum. I cant find the original post now.

Similar shortcuts exist for covariance and correlation coefficient. 

[MANOJ KUMAR Narayan]

great!!

[Satish]

These two useful

[Malabika Guha Mustafi]

@ CG,

Query on Variance/Standard Deviation formula for sample:30 Malabika Guha Mustafi > wrote: > >> @AC thank you for pointing out the formula. >> @ Sherlock thanks for the proof. >> So , sample variance = n/n-1 population

yes, I pointed out that n/n-1 population variance= sample variance 

Abhin  expanded with average formulation.

But, my question now whether this formula is still applicable for frequency data? Can we apply the same formula to compute s.d  in problem like 10 (mock test)?

[Malabika Guha Mustafi]

@ Tejasvi,

Please go through the link, easy way to get covariance

COVARIANCE SHORTCUT: Easy calculationscool. Thanks On Sunday, 15 November 2020 at 10:52:34 UTC+5:30 Satish wrote: On Sunday, November 15,

[CG]

@Malabika It can be used. I've shown how it can be used below.

Original Formula : SQRT [Sigma(x-xmean)^2 / N-1]

Computational Formula : SQRT [(Sigma(X2) - NxMean) / (N-1)]

Bikes Owned | Frequency

0 5

1 8

2 4

3 2

4 1

Using Computational Formula

Total number of bikes = 8 + 8 + 6 + 4 = 26

Total number of friends = 5 + 8 + 4 + 2 + 1 = 20

Mean (Bikes / Friends)= 26/20 = 1.3

Mean^2 = 1.69

x   = 0,0,0,0,0,1,1,1,1,1,1,1,1,2,2,2,2,3,3,4 

x^2 = 0,0,0,0,0,1,1,1,1,1,1,1,1,4,4,4,4,9,9,16

sigma (x)   = 26 

sigma (x^2) = 8+16+18+16 = 58

SD using computational formula = SQRT [(58 - 20x1.69)/19] = SQRT [1.27] = 1.1269

Using Original Formula

x-xmean = -1.3,-1.3,-1.3,-1.3,-1.3,-0.3,-0.3,-0.3,-0.3,-0.3,-0.3,-0.3,-0.3,0.7,0.7,0.7,0.7,1.7,1.7,2.7

(x-mean)^2 = 5x(1.3)^2+8x(0.3)^2+4x(0.7)^2+2x(1.7)^2+1x(2.7)^2

(x-mean)^2 = 5x1.69 + 8x0.09 + 4x 0.49 + 2x2.89 + 1x7.29 = 24.2

SD using original formula = SQRT [24.2/19] = SQRT [1.2736] = 1.1269

[Malabika Guha Mustafi]

Wow! 

So, for grouped data we just remember that sigma (x^2) = sigma ( fi X xi^2) { Actually for grouped data calculation xi^2 is replaced by fi X xi for each case}.

Thank you all of you. 
 This forum is awesome in true sense.

[Akshay Malik]

Another way!

Right now , I am not able to comprehend the formulae you guys are using. May be because my mind is just a bit tired right now!

But do tell me which of the two methods is easier!

[Malabika Guha Mustafi]

So, for computing sample deviation problem like 10.

We may approach like this way also,

mean=1.3

Square of mean= 1.69

N=20

Population variance= sigma (fi * xi^2)/N -(mean square)

Here, 

P.v= (58/20) -1.69 = 1.21

Sample variance= N/N-1 * population  Variance =20/19 * 1.21 =1.27

S. D of sample=root ( 1.27) = 1.1269

So, all formula is true for   both grouped and ungrouped data.

[Akshay Malik]

Okay! Both are almost the same formulae! Just a matter of putting it in a more concise way!

[Akshay Malik]

Yes. I agree.

On Sunday, November 15, 2020 at 1:19:11 PM UTC+5:30 Malabika Guha Mustafi wrote:

@ akshayam 

It depends on who is applying.

It is just like straight line  pollution problem ( refer Mock test maths). Some people approach from midnight to 10 a.m and some people approach from 8 p.m to 10 a.m.

Same thing just different approach.

 But, the formula CG  stated it is easier to remember.

Now with this formula, by multiplying N in different way we can get different formula presentation which are actually one.

[Sherlock Holmes]

Just a little correction in the notes, standard deviation after multiplication with a constant becomes mod(constant)* original standard deviation.(Not c squared).  

[Malabika Guha Mustafi]

To solve  the variance problem, how much time should be taken? Any idea?considering the short cut method. Thanks in advance.

[aishwary verma]

This is good work. Thanks man! 

Btw, as a standard of reference, IITM should be doing this, compressing what's taught in 4 weeks into a std. pdf will help everyone. 

Cheers!

[Tejasvi Hegde]

Thanks, will update 

[CG]

@Tejasvi

I made a PDF last night with the shortcut formulae. I've posted it to the group. Feel free to use it in your PDF. If you need the LaTeX for the equations so that you can add it your PDF also, then let me know.

[Tejasvi Hegde]

Thaknks @CG will update them 

[Tejasvi Hegde]

@CG can you please share word/excel doc its easy to copy from them for my notes

On Monday, November 16, 2020 at 9:54:11 AM UTC+5:30 CG wrote:

[Satya Mohapatra]

Good one 

SD multiply a constant will be C* SD and not Csquare * SD probably a typo.

On Friday, November 13, 2020 at 6:00:10 PM UTC+5:30 Tejasvi Hegde wrote:

Updated version 1.1

I have updated with ready reference of formulas in one place and chart of relation

On Friday, November 13, 2020 at 4:40:39 PM UTC+5:30 Sherlock Holmes wrote:

This is such a great work on your part.You have my sincere thanks for sharing it with us!!

On Friday, November 13, 2020 at 4:35:23 PM UTC+5:30 Mohan Palanisamy wrote:

Thanks Tajasvi. Great job.

by the by, Is SSE is related the std deviation.

[Satya Mohapatra]

Multiplying constant to variance also needs correction.

[Tejasvi Hegde]

IMPORTANT: There are couple of mistakes in my document, which I am yet to update,

But till then, please refer to the document prepared by CG on those points.

Its a nice short and very helpful document

https://groups.google.com/a/nptel.iitm.ac.in/g/ma1002-discuss/c/2yimKt1iip4

[Tejasvi Hegde]

And Honestly, I am lost in Maths and Stats as I have lost touch touch since decades. 

Memorizing them all and applying right one for right kind of real world questions , all in short of time...  Making me nervous and anxious.

Waiting anxiously for exam to get over with any result

On Monday, November 16, 2020 at 11:19:59 PM UTC+5:30 Tejasvi Hegde wrote:

Thank you all for suggestions.

Here is updated version 1.2

[Tejasvi Hegde]

Thank you all for suggestions.

Here is updated version 1.3

[rahul prakash]

Great Effort and thank you for sharing

Please include in a single table,  the change in the vales of Mode, Mean, Median, Variance, Std Dev upon adding and multiplying constants to values.  

[Prerna Dumka]

@Tejasvi @CG

Thank you for your generosity, these notes are really helpful.

Can you please suggest me a source for practicing ques in CT? It's my weak point so I'll be needing extra practice.

 

Thanks for your kind attention.

[Akshay Malik]

@ Prerna 

You may download an app to learn the fundamental concepts of programming.

I would recommend Programmming Hub: Learn to code.

Install it and learn the basics  of python language and practice a bit. Python is closest to the pseudocode. You will learn it in a day.

And you only need to understand the IF Block , WHILE LOOP,  the Assignment operators. Once you understand the logic, you can solve any question of CT.

 https://www.w3schools.com/python/python_conditions.asp

You can also go through this if you want to understand the IF STATEMENTS clearly

[Akshay Malik]

CT_Student = 'Prerna'

if (CT_Student undertstands LOGIC):

     CT_Sudent will be confident

else:

     CT_student may find it difficult to solve CT questions.

[Anand Iyer]

NameError: name ' LOGIC' not defined :)

[Akshay Malik]

Haha! True that. 

[Mohan Palanisamy]

Pls change the email of this group as below.

mohan...@gmail.com

Regards
Mohan

“DISCLAIMER: This e-mail contains privileged and/or confidential information and/or other information subject to legal restrictions and is meant for the sole use of the intended recipients. If you are not the intended recipient of this message or you have received the e-mail by mistake or whatsoever, please do not read, copy, use or disclose this message or its attachments. Please notify the sender immediately and destroy all copies of this message and any attachments. The views, opinions, and judgments expressed in this email are solely those of the author and same has not been reviewed or approved by SHAHI Exports Pvt. Ltd.. SHAHI Exports Pvt. Ltd. accepts no liability for the content of this email, or for the consequences of any actions taken on the basis of the information provided. WARNING: This e-mail message including attachment(s), if any, is believed to be free of any virus. However, it is the responsibility of the recipient to ensure for absence of viruses. SHAHI Exports Pvt. Ltd. shall not be held responsible nor does it accept any liability for any damage arising in any way from its use. Please take a note that if you enter into email correspondence with SHAHI Exports Pvt. Ltd., and/or any of its other subsidiaries, associate companies/ companies under the same management, this piece of communication may be subject to monitoring and storage in company archives.”

[Priyanshu Singh]

if i am applying variance formula given in ntoes updated 1.3 the computational formula ,

then ans for s.d of mcok q10 is comig as 1 wihch is not in range if coreect ans 1.1and 1.2?

[CG]

The answer to Mock Question 10 using both formulae was shown earlier. Here is the link - https://groups.google.com/a/nptel.iitm.ac.in/g/ma1002-discuss/c/yYjyIWX9-aQ/m/3ZsrA4q0AQAJ

[Cherian George]

Given below is the solution for tutorial 5 of week 3 using the shortcut formula.

On Wed, Nov 18, 2020 at 11:15 PM Malabika Guha Mustafi <malabi...@gmail.com> wrote:

@ CG or anyone, 

have you tried the formula  for Tutorial 5..week 3 ?

[Malabika Guha Mustafi]

Yes, Got it at the early morning :). Thanks for reply. Silly mistakes.

[Pragati Mungekar]

Yes, I have used all shortcut formulas and solved the Mock Test problems & most of the Tutorial problems (W3 Tut. 5) as well.

Thanks for that compilation of formulas.

Post Mock it helped me gain my confidence back in Stats.

Thanks,

Pragati.

On Wed, 18 Nov 2020 at 23:15, Malabika Guha Mustafi <malabi...@gmail.com> wrote:

@ CG or anyone, 

have you tried the formula  for Tutorial 5..week 3 ?

On Tuesday, 17 November 2020 at 12:50:40 UTC+5:30 CG wrote:

--
You received this message because you are subscribed to the Google Groups "Discussion forum for Statistics for Data Science I" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ma1002-discus...@nptel.iitm.ac.in.
To view this discussion on the web visit https://groups.google.com/a/nptel.iitm.ac.in/d/msgid/ma1002-discuss/627ca3c3-4571-4d90-96c4-1d5616e940dan%40nptel.iitm.ac.in.

[Malabika Guha Mustafi]

In statistics , this is now become a frequent issues. This subject is totally new for us at least for many of us. This conflict of basic fundamental things should be addressed by official announcement. If someone missed any thread ,he or she  is in wrong side.

for example. [Biserial  coefficient ]

stats1-...@onlinedegree.iitm.ac.in

18 Nov 2020, 17:10:01 (yesterday) 

to Discussion forum for Statistics for Data Science I, Tejasvi Hegde, yugesh maitreyan

Hello, 

You can use the sample standard deviation for calculating the point bi-serial coefficient, but in that case you need to use the sample correction in the formula. Instead of using sqrt(p0 *p1), use sqrt[(n0/n-1)(n1/n)].  But if you are using population standard deviation for calculation, then use sqrt(p0*p1). 

In tutorial 4, there is a mistake, we will send an announcement.

Also, please find the screenshot attached here for the formula.

Thanks

Nikita

Statistics course support team

This type of things are addressed only when someone raised the issue and that is also in a thread. 

Team, Please address such issues through official announcement  not in thread.( Resource from Instructor  or email can be used, but not in forum. It is not possible for us to follow each thread in forum or follow other's doubt clearing session).

Thanks in advance.

I posted in other threads also. It is really disturbing.

[soniya duge]

Yes agree to this as I too got to know from this thread itself.

Also thanks everybody who has contributed to the compilation of all the shortcut formulas.

It really helped me to revise everything fast and crear doubts.

Thanks & Regards,

Soniya Duge

To view this discussion on the web visit https://groups.google.com/a/nptel.iitm.ac.in/d/msgid/ma1002-discuss/dc605f61-5ac0-475d-b320-a449f78ed734n%40nptel.iitm.ac.in.

[soniya duge]

Anybody have similar notes for Maths to share? Please...

Thanks & Regards,

Soniya Duge

[Akshay Malik]

I have digital - handwritten notes. But I made them for my own convenience and cannot guarantee absence of typos.

[soniya duge]

np. I can follow those. Can you share them please?

Thanks & Regards,

Soniya Duge

[Malabika Guha Mustafi]

@ sonny, Tesajvi compiled beautiful notes on maths. Please search it. It may be helpful to  you. Akshayam also made  good notes. I forgot the name, One person shared a beutiful book on set , relation, function. I reposted that in separate link. Everything you 'll get in maths forum.

[soniya duge]

Thanks Malabika!!! will check them in the Maths Forum.

[Akshay Malik]

@Soniya

I have shared my One Note  Notebook with you!

[Akshay Malik]

@Soniya 

Check your mail and see if it's working.

[Sanchi Priyanshu]

i  have arranged all formula ,units and effects of outliers for last minute revision,can anyone of the experts here please check this one page of all formulae units and affects i have written and see if it is correct or not,please rectify me if anything is wrong

[Akshay Malik]

@Sanchi Update the Point-BiSerial formula by writing that Sx stands for Population Standard Deviation

Also add the shortcut formulas in your list

[Akshay Malik]

Also When you multiply all elements with C

new variance = C^2  * Old Variance

[Akshay Malik]

Variance is not affected by adding a constant

[Anand Iyer]

Sanchi,

look at the last table.  There are a few mistakes...

You received this message because you are subscribed to a topic in the Google Groups "Discussion forum for Statistics for Data Science I" group.
To unsubscribe from this topic, visit https://groups.google.com/a/nptel.iitm.ac.in/d/topic/ma1002-discuss/yYjyIWX9-aQ/unsubscribe.
To unsubscribe from this group and all its topics, send an email to ma1002-discus...@nptel.iitm.ac.in.
To view this discussion on the web visit https://groups.google.com/a/nptel.iitm.ac.in/d/msgid/ma1002-discuss/395c74aa-694c-4736-8e86-a467775d6bc2n%40nptel.iitm.ac.in.

--

Cheers,

Anand

[Akshay Malik]

new Covariance = C^2 * Covariance

And Similar for Correlation

[Dhiman Sengupta]

Just to double check with you all..

Adding a constant- change in mean, median, mode , no change in range, variance,  covariance, std deviation, 

Multiple a constant: change in mean, median, mode, range, variance , co variance,  std dev

Outlier : impact on mean, variance, std dev and correlation coefficient, R^2 but no impact on mode, median , IQR

On Thu, 19 Nov 2020, 11:52 am Akshay Malik, <akshaym...@gmail.com> wrote:

Variance is not affected by adding a constant

--
You received this message because you are subscribed to the Google Groups "Discussion forum for Statistics for Data Science I" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ma1002-discus...@nptel.iitm.ac.in.

To view this discussion on the web visit https://groups.google.com/a/nptel.iitm.ac.in/d/msgid/ma1002-discuss/1dfbedd9-a150-4475-ad0e-d160c8fe7650n%40nptel.iitm.ac.in.

[Malabika Guha Mustafi]

Satish highlighted the same thing in a tabular form .it 's quite handy . It is in this thread only.

[Dhiman Sengupta]

I found most of the conversations but unfortunately not the desired one. Can you please repost if possible.. 

On Thu, Nov 19, 2020 at 12:21 PM Malabika Guha Mustafi <malabi...@gmail.com> wrote:

Satish highteled the same thing in a tabular form .it 's quite handy . It is in this thread only.

--

Thanks,

Dhiman Sengupta

Bangalore

Ph: +91 7338130710

[Malabika Guha Mustafi]

[Malabika Guha Mustafi]

[Dhiman Sengupta]

Thank you so much Malabika..

[Malabika Guha Mustafi]

@ Sathis, I have seen  you answering regarding to the outlier in different threads. Is it possible for you to post it in a tabular form.

[_Music_ Viking]

In that reference pdf, is the formula given to calculate SSE correct?

[Akshay Malik]

https://groups.google.com/a/nptel.iitm.ac.in/g/ma1002-discuss/c/P5pho3K7qzw

Updated pdf

[Priyanshu Singh]

if u can see that corelation doesnt change on adding or multiplying constant,so  i think one of u  might be wrong,please recify and reply anyone,malabika? 

[Malabika Guha Mustafi]

@priyansu

In denominator standard deviations' c  will cancelled the  covariance c..so no effect on correlation.

[akash shrivastava]

Rating is ordinal or interval scale ? In lectures slide it is mentioned ordinal in faq it is interval 

On Fri 13 Nov, 2020, 6:00 PM Tejasvi Hegde, <tejasv...@gmail.com> wrote:

Updated version 1.1

I have updated with ready reference of formulas in one place and chart of relation

On Friday, November 13, 2020 at 4:40:39 PM UTC+5:30 Sherlock Holmes wrote:

This is such a great work on your part.You have my sincere thanks for sharing it with us!!

On Friday, November 13, 2020 at 4:35:23 PM UTC+5:30 Mohan Palanisamy wrote:

Thanks Tajasvi. Great job.

by the by, Is SSE is related the std deviation.

On Friday, November 13, 2020 at 3:21:21 AM UTC+5:30 Tejasvi Hegde wrote:

Here is stats notes I had prepared for myself. Thought it may be helpful to many

This basically contains highlights and ready reference

 

For Maths notes, check 

https://groups.google.com/a/nptel.iitm.ac.in/g/ma1001-discuss/c/ubIcoEBawtg

ps: Let me know in case of any error  

--

You received this message because you are subscribed to the Google Groups "Discussion forum for Statistics for Data Science I" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ma1002-discus...@nptel.iitm.ac.in.

To view this discussion on the web visit https://groups.google.com/a/nptel.iitm.ac.in/d/msgid/ma1002-discuss/f81b1d42-518a-42e5-9606-bb408b12eacen%40nptel.iitm.ac.in.

[Akshay Malik]

Thanks @priyanshu for pointing out my mistake.

Types of Rating :

1. Bad, Average, Good --- Ordinal Scale

2. 1 , 2, 3, 4, 5 ---- Interval Scale

3. 0, 1, 2, 3, 4, 5 ---- Ratio Scale

3. Average of all the ratings as we see on the playstore ---- Ratio scale

[Anand Iyer]

Can be either.  

If the rating occurs in 1 - 5 (1, 2, 3,4,5), then the values are numerical and hence scale is interval.

If the rating occurs in discrete, categorical quantities, then the scale is ordinal.  For example, Good-Bad-Ugly or Very satisfied-Satisfied-Not Satisfied.

Now, if the rating occurs in 0-5, then the scale becomes ratio since there's an absolute zero.  Again, this is debatable, because the interpretation of the 0 to be absolute, depends on the context.

You received this message because you are subscribed to a topic in the Google Groups "Discussion forum for Statistics for Data Science I" group.
To unsubscribe from this topic, visit https://groups.google.com/a/nptel.iitm.ac.in/d/topic/ma1002-discuss/yYjyIWX9-aQ/unsubscribe.
To unsubscribe from this group and all its topics, send an email to ma1002-discus...@nptel.iitm.ac.in.
To view this discussion on the web visit https://groups.google.com/a/nptel.iitm.ac.in/d/msgid/ma1002-discuss/CAHRuO6OVOLXCgwnhjK%3DX6qOXcZRWgguy5PA3DfDKQyVv0t51Qw%40mail.gmail.com.

--

Cheers,

Anand

[Kaushik Deka]

If x(i) = c1*x(i) , y(i) = c2 * y(i) , c1 NOT= c2,

=> New r = (c1*c2 / |c1*c2|) * Old r

So, when c1, c2  have same sign => New r = Old r

But, when c1, c2 have opposite signs => New r = - (Old r) i.e. the regression line will be mirror image, of previous case, on X-axis or Y-axis!

One should consider all cases in deductions. :)

Regards,

Kaushik

--
You received this message because you are subscribed to the Google Groups "Discussion forum for Statistics for Data Science I" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ma1002-discus...@nptel.iitm.ac.in.

To view this discussion on the web visit https://groups.google.com/a/nptel.iitm.ac.in/d/msgid/ma1002-discuss/CAJs2mJKdDmyNnZy-N1ndz%3Doi3svdkMGcaB4bBf7MsMok4ShLmQ%40mail.gmail.com.

[Malabika Guha Mustafi]

Thank you for the clarification. 

[Kaushik Deka]

Can the average rating on playstore take 0 (star) ? :)

[Pranav Dave]

So it has an absolute zero(not rating an app is similar to it having 0)? Now do all ratio scales have an infinite range?

[Kaushik Deka]

Hi Pranav,

It was a rhetorical question. I meant since playstore never accepts 0 star as a rating ( i.e. no sample point with 0 ), so its average rating can never be 0. Therefore, its scale of measurement is interval, and NOT ratio, contrary to what Akshay has mentioned above.

[Pranav Dave]

I get that your question was rhetorical but mines a genuine doubt. When we say height is zero or weight is zero we mean that we aren't measuring anything right. So you could also have 0 stars(unrated). Now all ratio scales of measurement seem to have an infinite range which is why I am ruling rating out as ratio. 

[Kaushik Deka]

Height, weight have ratio scale because absolute 0 is meaningful when you measure them in the SI units. You can record these values as relevant data points. 0 star in playstore is meaningless because you CANNOT give a 0 rating - minium is 1 star, right? Not rating is NOT= giving 0 rating; so, no data point with 0 rating! I'm not sure what you mean by "ratio scales of measurement seem to have an infinite range", but infinity or continuity has nothing to do with the definition of a ratio scale. 

[Sanchi Priyanshu]

i think covariance is wrong in pdf formula given by cg

as covariance multiplied by constant gives c * covairance and not c^2

[Ashish bharadwaj]

Thanks a lot for updated(1.3) version notes .

[CG]

If you look at rule 6 in your attached image
 COV(cX,kY) = ck. COV(X,Y)

 COV(cX,cY) = c.c. COV(X,Y) = c^2 . COV(X,Y)

So C^2 * COV(X,Y) is correct when every x and y are multiplied by C

On Saturday, November 21, 2020 at 1:09:17 PM UTC+5:30 Sanchi Priyanshu wrote:

[Aditi Prasad]

Can somebody explain the difference between cross-sectional and time series data??  

To view this discussion on the web visit https://groups.google.com/a/nptel.iitm.ac.in/d/msgid/ma1002-discuss/db7702c9-a700-42c6-82af-d85da27fe97en%40nptel.iitm.ac.in.

[Tejasvi Hegde]

Time Series is the data along time line 

e.g. Temperature reading taken in a city at different time or date  i.e. data taken at 10am, 11am etc...  

Cross section is the data taken at same time

e.g. Temperature reading in multiple cities at same time. i.e. temperature at 10am in City1, City2, City3 etc.

Hope its clear now ?

[Gada Electronics]

Hi,

Can you please explain what is Boolean, String, Character if possible?

Thanks in advance

On Tuesday, 17 November 2020 at 09:38:40 UTC+5:30 akshaym...@gmail.com wrote:

@ Prerna 

You may download an app to learn the fundamental concepts of programming.

I would recommend Programmming Hub: Learn to code.

Install it and learn the basics  of python language and practice a bit. Python is closest to the pseudocode. You will learn it in a day.

And you only need to understand the IF Block , WHILE LOOP,  the Assignment operators. Once you understand the logic, you can solve any question of CT.

 https://www.w3schools.com/python/python_conditions.asp

You can also go through this if you want to understand the IF STATEMENTS clearly

On Tuesday, November 17, 2020 at 8:18:14 AM UTC+5:30 prerna...@gmail.com wrote:

@Tejasvi @CG

Thank you for your generosity, these notes are really helpful.

Can you please suggest me a source for practicing ques in CT? It's my weak point so I'll be needing extra practice.

 

Thanks for your kind attention.

On Tuesday, 17 November 2020 at 01:54:00 UTC+5:30 Tejasvi Hegde wrote:

Thank you all for suggestions.

Here is updated version 1.3

On Monday, November 16, 2020 at 11:25:02 PM UTC+5:30 Tejasvi Hegde wrote:

And Honestly, I am lost in Maths and Stats as I have lost touch touch since decades. 

Memorizing them all and applying right one for right kind of real world questions , all in short of time...  Making me nervous and anxious.

Waiting anxiously for exam to get over with any result

On Monday, November 16, 2020 at 11:19:59 PM UTC+5:30 Tejasvi Hegde wrote:

Thank you all for suggestions.

Here is updated version 1.2

On Monday, November 16, 2020 at 8:57:38 PM UTC+5:30 Tejasvi Hegde wrote:

IMPORTANT: There are couple of mistakes in my document, which I am yet to update,

But till then, please refer to the document prepared by CG on those points.

Its a nice short and very helpful document

https://groups.google.com/a/nptel.iitm.ac.in/g/ma1002-discuss/c/2yimKt1iip4

On Monday, November 16, 2020 at 2:05:29 PM UTC+5:30 Satya Mohapatra wrote:

Multiplying constant to variance also needs correction.

On Monday, November 16, 2020 at 12:51:43 PM UTC+5:30 Satya Mohapatra wrote:

Good one 

SD multiply a constant will be C* SD and not Csquare * SD probably a typo.

[Tejasvi Hegde]

Boolean is Logical value , i.e. true or false. (some times 0 for false, 1 for true is also considered)

Character   : Represents single character (as an alphabet in English language) even digits (when considered in non numerical operation context), special characters, space etc.  i.e. a, A, z, Q, - , 0 , 2 etc.   

String: When characters are combined/bound together they form string (similar to sentences in English language) i.e. APz34 is a string

Character is usually represented in single quote i.e. 'A' , 'z' etc.

String is represented within double quote (in some languages, single quote is also used) i.e. "APz34 12"   (recognize space character in this string)

Hope this is clear ?