Amelia McNamara

Пікірлер

@aatikakanwal5873 4 күн бұрын

what is meant by lag in simplest terms?

@GirijaPrasadPatnaik 2 ай бұрын

Thank you so much

@reactionchamber 3 ай бұрын

I really like the idea of the "allows us to say" sentences. Thank you!

@gaurav012 6 ай бұрын

Thank you! I think I am finally starting to understand the difference between confidence and prediction intervals!

@FranciscoMedina-p6c 7 ай бұрын

Thanks a lot, your recomendatio about tinytex, came for me exactly in the moment that we needed. We were in a RStudio Course. Blessings!

@bkarim7349 9 ай бұрын

Great video

@bkarim7349 9 ай бұрын

Thank you. Very usefule

@Francisco-j1e 9 ай бұрын

pairs(DF[,VQ], col=DF$VN, pch=22,col=DF$VN) ... this one is a pairs with numerical (VQ) vs a categorical (VN) and shows the basis as the normal pairs but this one has colors i suppose thats where it relates to the VQ variable, but i cant interpret the colors....

@VinayakPattanashetti 10 ай бұрын

best

@bingbro21 11 ай бұрын

Good morning Professor, huge thanks for creating these videos for us. They are extremely helpful. I was wondering if you've covered scaled power transformations in later chapters? I found it quite difficult to understand. Thnak you very much !

@dasrotrad Жыл бұрын

This did not work for me either. ChatGPT suggested I uninstall and reinstall .. then ChatGPT gave up. tinytex::uninstall_tinytex() # Re-Install TinyTeX together with prebuilt() tinytex::install_tinytex() # Install TinyTeX packages, including xelatex tinytex:::install_prebuilt() I kept getting the error: running xelatex - 1 command xelatex not found, attempting install finding package for xelatex > 1 package to install > installing xelatex (1 of 1) ERROR: tlmgr returned a non zero status code The system cannot find the path specified. ChatGPT checked my xelatex path and said it was there and I should post on Posit Community.

@naufalfarabbi3323 Жыл бұрын

I have problem, when i install tinytex, the command stop at Running fc-cache -v -r > It didn't show any installation progress nor message of installation successfull (i wait 15 minute ,my machine is not slow pc)

@AmeliaMcNamara Жыл бұрын

Hmm, is that when you're trying to install it at first, or when you're testing it? I've seen the first time running tinytex to take a really long time (maybe as much as 15 minutes or more) because it's installing a lot of packages.

@enockbanda879 Жыл бұрын

Very Helpful Video. Thank you

@dmitriidoroftei4478 Жыл бұрын

Very helpful, thank you!

@yevgenympinyuri6983 Жыл бұрын

Thank you for the video it really helped

@keunohkim3642 Жыл бұрын

nice

@zainknbehram4225 Жыл бұрын

i have time series data and want to be forecasting on it by different models as i.e ANN Model , Neural Network , etc so are you help me with this work

@itdoliva Жыл бұрын

Loved it! Thanks for creating and publishing it <3

@AnArtisticToad Жыл бұрын

My class is also using stat key, so I needed a video going over it this way. Thanks for explaining it multiple ways and including the summaries.

@algorithmo134 Жыл бұрын

Hi Professor, at 4:05 what does SSM mean and what does RSS mean and why is SSE full in the last row?

@AmeliaMcNamara Жыл бұрын

SSM is the sum of squares for the model. Since we're comparing two models, we have the SSM for the full model and SSM for the reduced. RSS is the residual sum of squares, otherwise known as the SSE or sum of squares of the error. SSE full is the sum of squares of the errors from the full model. I've organized my models in increasing order of complexity (small to large) so the full model is the one at the bottom.

@algorithmo134 Жыл бұрын

@@AmeliaMcNamara At 3:51, I see that the value of 49.704 is found by subtracting the RSS from model 1 and model 2 but the formula presented is SSM full - SSM reduced in the numerator. I dont see how since neither model 1 nor model is the full model so can you please explain how do you get (SSM full -SSM reduced) = 49.704?

@hydrologicmodeling9020 Жыл бұрын

Thank you for a clear video!

@alessandrorosati969 Жыл бұрын

How is it possible to generate outliers uniformly in the p-parallelotope defined by the coordinate-wise maxima and minima of the ‘regular’ observations in R?

@theJasonLee 2 жыл бұрын

Data scientist and graduate student in CS at UIUC checking in here. Just wanted to thank you for this clear video. It is hard to come across such clear videos with R! Thank you!

@ketchup7108 2 жыл бұрын

You are the best

@NeoTokyo-Akira 2 жыл бұрын

Hello professor, may I know which book for the class

@AmeliaMcNamara 2 жыл бұрын

I'm referring to the sections of the "Lock 5" book, Statistics: Unlocking the Power of Data, by Robin H. Lock, Patti Frazer Lock, Kari Lock Morgan, Eric F. Lock, Dennis F. Lock. I was using the second edition when I recorded the videos, but the third edition is out now.

@danielebaldoni2181 2 жыл бұрын

Very clear the explanation. How would be the formula for the standard error of a prediction interval? Yhat +/- t* ..... .? Could i use it to predict an interval of time series?

@AmeliaMcNamara 2 жыл бұрын

The equation for the standard error for the prediction interval is on the screen around 5:28. As for time series, they have much wider prediction intervals! Check out the end of kzbin.info/www/bejne/rF7SeX1oZ7mmesU for an introduction.

@KN-tx7sd 2 жыл бұрын

thank you, excellent. It's a new package (leaps) and versatile. From here can we pipe this information for prediction as well as generating ROC and AUC with bootstrapping. If there are any good packages that could be used with 'leaps' package please let us know, Another question is if you have to say around 100 variables and you want to select the top 10 or 20 best predictors can this package will be able to do it. Thank you again.

@ahmed007Jaber 2 жыл бұрын

thank you for this; how would you add subtotals by group and grand total while displaying all observations just like what you would do in excel?

@AmeliaMcNamara 2 жыл бұрын

I'm showing the "tidyverse" way of doing summary statistics. In the tidyverse, grand totals don't make sense, because of the way "tidy data" is defined. If you want something with grand totals, try the adorn_ functions from the {janitor} package, sfirke.github.io/janitor/articles/tabyls.html#the-adorn-functions

@aymanrafif5904 2 жыл бұрын

Awesome! Can you post the markdown in rpubs please?

@vamsinvizag4936 2 жыл бұрын

Thank you. This was very helpful!

@shah_f1 2 жыл бұрын

the steps u did about R_profile for reticulate can also be done for Quarto. just add 1 more line in the .Rprofile as QUARTO_PYTHON: "-------" where --- is the path to ur miniconda May be it will ask u about jupyter installation which can be done in miniconda

@AmeliaMcNamara 2 жыл бұрын

Thanks! Yeah I realized once I restarted RStudio that I could get Quarto going with Python chunks. It did prompt me about conda install jupyter.

@truongphu7407 2 жыл бұрын

Thanks for your helpful tutorial. Could you tell how to run LSD test for multi-variables data set? If I run LSD test one by one, tit will takes long time due to there are only four groups but 15 variables.

@duanebarrosfonseca 2 жыл бұрын

Hi, can I use the same approach for confidence and prediction intervals of nonlinear functions?

@AmeliaMcNamara 2 жыл бұрын

The interpretation sentences should be about the same, but the standard error calculations may be different, depending on the model.

@ee_li 2 жыл бұрын

Thank you--I found this clear and easy to follow. Keep up the great work!

@sgssgssgs 2 жыл бұрын

Interesting video. I first came across the National Atlas of the US in the mid-1980s at the NY Public Library Map Room (a great place, BTW). After spending time with it, I had to own one! I looked to buy one for many years, but could never find one for sale. Then, eBay came along! On eBay, they often brought low-ish prices, so I ended up with 5 or 6 of them, which I still have. They are often ex-library, since they cost $100 in 1970--a pretty penny back then, only institutions bought them. These days, with everything being digital, they mostly have all been deaccessioned. (Not to mention that all the thematic data are now obsolete.) I have since become a collector of National Atlases.

@AmeliaMcNamara 2 жыл бұрын

Amazing! Yes, I bought mine on eBay as a gift to myself when I finished my PhD.

@jakeslaton2849 2 жыл бұрын

Really wish I would have found these videos before completing my master's thesis. Thanks for sharing!

@AmeliaMcNamara 2 жыл бұрын

You're very welcome!

@claragranell 2 жыл бұрын

Great video! You mentioned in the video that there was an updated version of the Cleveland & McGill 1984 paper. Do you have the reference for it? Thanks!

@AmeliaMcNamara 2 жыл бұрын

I think I was referring to the Heer and Bostock paper from 2010, which I discuss more here: kzbin.info/www/bejne/h33MgaBuj8t1m8U

@claragranell 2 жыл бұрын

@@AmeliaMcNamara thank you for the reference! I'll check it out and your video as well!

@notterboutuyer2781 2 жыл бұрын

should not we add "on average" in the end of statements we make from the models?

@AmeliaMcNamara 2 жыл бұрын

You definitely want something to "hedge" your statement, so that you are capturing that it's just what the model predicts, not reality. I like "we expect" as that language but others may prefer "on average." To me they do the same sort of thing.

@notterboutuyer2781 2 жыл бұрын

@@AmeliaMcNamara thank you for the response. i really like your way of teaching. learning a lot. can you please share which book you follow at stat 320. i like to follow the videos along with the book. thanks again for the great content.

@AmeliaMcNamara 2 жыл бұрын

Thanks for your kind words! I use the Stat2 book by Ann Cannon, George Cobb, et al. With my students I have switched to the 2nd edition, but the 1st edition is pretty similar.

@kunalbali810 2 жыл бұрын

Have you used NetCDF files to plot spatial maps? for example, anomaly, climatology, etc spatial plots using R?

@AmeliaMcNamara 2 жыл бұрын

I have not! I love maps, but when I work with them in R I have used shapefiles with both the sp and sf packages (see some code here: www.amelia.mn/sds236/MakingMaps.html). I'll have to look into NetCDF.

@ojassaxena2891 2 жыл бұрын

Can we make observations with help of anova table??

@apurwasharma3582 2 жыл бұрын

Please post more such videos. I was really amazed by this video.

@kelumsenaka4146 2 жыл бұрын

Good explanation!!! Can you attach the code file to the description.

@Monkeysal07 2 жыл бұрын

Hello, does the continuous measurement of temperature over time of a stream of liquid constitute a violation of the independence condition? Thank you

@AmeliaMcNamara 2 жыл бұрын

Yes, I would consider that to be a violation of the condition. Knowing the temperature a second ago will give you some information about the temperature right now. Time series analysis would be more appropriate than linear regression, unless the relationship is very linear.