- 47
- 429 878
StatDOE by Rosane Rech
Canada
เข้าร่วมเมื่อ 27 พ.ค. 2020
Hi there, my name is Rosane Rech.
I am a Chemical Engineer with a Ph.D. in Molecular and Cellular Biology. During the last 25 years, I have been developing research on bioprocess, food and chemical engineering. During this time, I became passionate about going over data, statistical analysis, and data visualisation.
Recently, I have learned the statistical programming language R and became a fan of its potential for statistical analysis applied to the design of experiments and data visualisation.
This site was created to share valuable materials on statistics, design of experiments and data visualisation.
The material and tutorials on R published here follow a different approach from most R tutorials. Instead of tackling specific functions, the tutorials focus on covering the complete analysis of a data set and on the step-by-step of building high-quality plots, graphs, and charts.
I hope you enjoy the materials!
I am a Chemical Engineer with a Ph.D. in Molecular and Cellular Biology. During the last 25 years, I have been developing research on bioprocess, food and chemical engineering. During this time, I became passionate about going over data, statistical analysis, and data visualisation.
Recently, I have learned the statistical programming language R and became a fan of its potential for statistical analysis applied to the design of experiments and data visualisation.
This site was created to share valuable materials on statistics, design of experiments and data visualisation.
The material and tutorials on R published here follow a different approach from most R tutorials. Instead of tackling specific functions, the tutorials focus on covering the complete analysis of a data set and on the step-by-step of building high-quality plots, graphs, and charts.
I hope you enjoy the materials!
How to calculate the real-world values of the variable in central composite designs.
This video shows the relationship between the natural and coded variables for Central Composite Designs / Response Surface / Design of Experiments / DOE
statdoe.com/courses/
statdoe.com/courses/
มุมมอง: 490
วีดีโอ
Design of Experiments for Mixtures
มุมมอง 55210 หลายเดือนก่อน
This video is a sample lesson of the course "Designs of Experiments for Mixtures" statdoe.com/MixtureDesigns
RSM 08 Desirable features for Experimental Designs
มุมมอง 2Kปีที่แล้ว
Response Surface Methodology (RSM) course - Lesson 8/8 statdoe.com/rsm/
RSM 07 Analyzing a Central Composite Design in R
มุมมอง 3.9Kปีที่แล้ว
Response Surface Methodology (RSM) course - Lesson 7/8 R tutorial statdoe.com/rsm/ www.statdoe.com
RSM 06 Central Composite Design
มุมมอง 5Kปีที่แล้ว
Response Surface Methodology (RSM) course - Lesson 6/8 statdoe.com/rsm/ . www.statdoe.com
RSM 05 Designs with Central Points - Interpreting a significant lack-of-fit
มุมมอง 2.7Kปีที่แล้ว
Response Surface Methodology (RSM) course - Lesson 5/8 statdoe.com/rsm/ . www.statdoe.com
RSM 04 Method of the Steepest Ascent
มุมมอง 4.3Kปีที่แล้ว
Response Surface Methodology (RSM) course - Lesson 4/8 statdoe.com/rsm/ . www.statdoe.com
RSM 03 Analyzing a 2ˆ2 Factorial Design with Central Points - R tutorial
มุมมอง 4.9Kปีที่แล้ว
Response Surface Methodology (RSM) course - Lesson 3/8 statdoe.com/rsm/ . www.statdoe.com
RSM 02 The starting design: 2ˆ2 Factorial Designs with Central Points
มุมมอง 6Kปีที่แล้ว
Response Surface Methodology (RSM) course - Lesson 2/8 statdoe.com/rsm/ . www.statdoe.com
RSM 01 Introduction to Response Surface Methodology
มุมมอง 10Kปีที่แล้ว
Response Surface Methodology (RSM) course - Lesson 1/8 statdoe.com/rsm/ . www.statdoe.com
Building a Pie-Donut Chart in R
มุมมอง 9K3 ปีที่แล้ว
R code: statdoe.com/pie-donut-chart-in-r/ packages: ggplot2, webr, dplyr functions: PieDonut . www.statdoe.com
Add and position letters indicating significant differences in bar plots
มุมมอง 19K3 ปีที่แล้ว
R code: statdoe.com/cld-customisation/ packages: ggplot2, ggthemes, multcompView, dplyr functions: aov, TukeyHSD, multcompLetters4, group_by, summarise, arrange, ggplot, geom_bar, geom_errorbar, geom_text, ylim, theme_few, vjust, just . www.statdoe.com
Desirability functions
มุมมอง 4K3 ปีที่แล้ว
Design of Experiments for Optimisation Enrol Now: statdoe.com/doeopt Website: statdoe.com Courses: statdoe.com/courses . www.statdoe.com
Barplots for the Results of a Three-Factor Experiment
มุมมอง 8K3 ปีที่แล้ว
R code: statdoe.com/barplots-for-three-factors/ This video presents a suggestion on how to present the results of a three-factor experiment using bar plots. The code can be accessed at: statdoe.com/barplots-for-three-factors/ R packages: ggplot2, ggthemes, dplyr, multcompview, egg R functions: aov, TukeyHSD, mlupcompLetters4, group_by, summarise, arrange, ggplot, geom_bar, geom_errorbar, xlab, ...
Boxplot with Letters Indicating Significant Differences
มุมมอง 42K3 ปีที่แล้ว
One-Way ANOVA, Tukey’s test, Box Plot with ggplot R code: statdoe.com/one-way-anova-and-box-plot-in-r/ Courses: statdoe.com/courses/ 0:00 Introduction 0:46 Description of the data set 1:01 Loading the libraries and the data set 1:55 Analysis of Variance - ANOVA 2:30 Tukey's test 3:10 Letters do indicate significant differences (cdl) 3:50 Table with mean, third quantile and cdl 5:05 Basic Boxplo...
Publication-Quality Scatterplots for Two Factors with ggplot - Two-Way ANOVA with R - tutorial 4
มุมมอง 3.9K3 ปีที่แล้ว
Publication-Quality Scatterplots for Two Factors with ggplot - Two-Way ANOVA with R - tutorial 4
Publication-Quality Barplots for Two Factors with ggplot - Two-Way ANOVA with R - tutorial 3
มุมมอง 13K3 ปีที่แล้ว
Publication-Quality Barplots for Two Factors with ggplot - Two-Way ANOVA with R - tutorial 3
Tukey's Test and Letters to Indicate Significant Differences - Two-Way ANOVA with R - tutorial 2
มุมมอง 29K3 ปีที่แล้ว
Tukey's Test and Letters to Indicate Significant Differences - Two-Way ANOVA with R - tutorial 2
Two-Way ANOVA with R - Analysis of Variance
มุมมอง 19K3 ปีที่แล้ว
Two-Way ANOVA with R - Analysis of Variance
[COURSE] Design of Experiments for Optimisation
มุมมอง 1.9K3 ปีที่แล้ว
[COURSE] Design of Experiments for Optimisation
One Way ANOVA 4 - Bar Plots Coloured According Tukey's Test using ggplot
มุมมอง 10K3 ปีที่แล้ว
One Way ANOVA 4 - Bar Plots Coloured According Tukey's Test using ggplot
One-Way ANOVA 3 - Using Superscript Letters in Tukey's Test
มุมมอง 49K3 ปีที่แล้ว
One-Way ANOVA 3 - Using Superscript Letters in Tukey's Test
One-Way ANOVA R tutorial 2 - Tukey's Test
มุมมอง 14K3 ปีที่แล้ว
One-Way ANOVA R tutorial 2 - Tukey's Test
One-Way ANOVA R-tutorial 1 - Analysis of Variance
มุมมอง 17K3 ปีที่แล้ว
One-Way ANOVA R-tutorial 1 - Analysis of Variance
DoE 67: Final Visualization of the Results - ggplot package
มุมมอง 1K3 ปีที่แล้ว
DoE 67: Final Visualization of the Results - ggplot package
DoE 64: Building Fractional Designs in R - FrF2 package
มุมมอง 7K3 ปีที่แล้ว
DoE 64: Building Fractional Designs in R - FrF2 package
DoE 63: One-Quarter Fraction of the 2ˆk Designs
มุมมอง 9043 ปีที่แล้ว
DoE 63: One-Quarter Fraction of the 2ˆk Designs
DoE 32: Analysing a 2^2 fatorial design in R
มุมมอง 8K3 ปีที่แล้ว
DoE 32: Analysing a 2^2 fatorial design in R
This is clear and easy to follow. superb!
I would only add, the specific value of alpha / position of axial points depends on k (number of variables). It tends to increase as you add more variables.
this was what I needed!
Thanks🎉
Hickle Plaza
Fantastic tutorial, thank you! But I'm new to R, how can I download the data and code files into RStudio? I tried copy-pasting the code on your website, but it got stuck at library(rsm) because "there is no package called rsm". Do I need to download an RSM package separately? Thanks!
Fixed with a little google search! For anyone else with the same problem, install.packages("rsm"). That seemed to work!
Hey! If I run your code, there is an Error after "Tk$cld <- cld$Letters", saying "Error in `$<-.data.frame`(`*tmp*`, cld, value = c("a", "a", "ab", "b", : replacement has 6 rows, data has 1"
Hello Rosane....! I was your student at Udemy... for the DOE for Optimization course... All material you present is excellent... Thanks for your help.
hello have u join this paid course
@@meghamane2184 Yes I did. It was excellent. And just what I needed to start understanding R language. In special for my job doing DoE. You need to code a little but it is fan to learn.
I am so confused. How do you tell the significance? Is it by seeing if they are in close range or maybe far apart or what
Very informative
how do we get the resulting yield, or yhe percentage?
it depends what u are looking for. For exemple, if you are looking for the optimation condition to cultive some bacteria, maybe, you will measure the UFC (colony forming unit) as the response. So you want to find the point when UFC is maximus, or i could look to the optimation condition to have bacterial spores, so i can put the response as the % of spores done when he look to the vegetative cells ( % spores formation = (spores/vegative cell)*100
Interesting video. How can I codify X1 if each point has a logarithmic increment? For example, A = 0, 1, 10, 100. x1 ~ (A - 100)/Log(10), is it correct? Thanks
Thanks som i find, your tutorial interesting, please can you help me out with the question below? Q. Write on Augmentation of 2k factorial experiment with center point to handle curvature and lack of fit
Hello! Thansks so much! I have a question: In 3:34, with 16 experiments you code suggest a 4 resolution instead of a 5 resolution: FrF2(nfactors = 5, resolution = 4) Which one is correct?
Can this be used for single replication? 😢
Yes, if you have enough degrees of freedom.
Muito origado pelo vídeo! Ajudou demais a entender essa confusão de letra.
Great video, thank you so much!
I ended up with a different ANOVA table below. Can someone tell me where I made a mistake in my R code? Df Sum Sq Mean Sq F value Pr(>F) Glass 2 267 133 0.888 0.429 Temp_Factor 2 3141469 1570734 10450.918 <2e-16 *** Glass:Temp_Factor 4 534 133 0.888 0.491 Residuals 18 2705 150
getwd() setwd("D:/") library(readxl) R <- read_excel("D:/R.xlsx") View(R) library(dplyr) # Assuming your dataframe is named excel_data data_summary <- R %>% group_by(Glass, Temp) %>% summarise(mean = mean(`Light`, na.rm = TRUE), sd = sd(`Light`, na.rm = TRUE)) %>% arrange(desc(mean)) # View the data_summary dataframe View(data_summary)
thank a lot a beatifull skills from beatifull Lady!
nice :)
Thanks a lot! This tutorial is easy to follow
At 3:56, assuming 0.71 and 1.4 were not significantly different, how do I draw the line? Do I draw it from 0.71 all the way to 1.4 or do I skip 1.02? The video was amazingly helpful, by the way. Thank you!
Hi! What code should I input to get CDL if I did Dunn’s Test? Thanks!
Direction of steepest ascent is just the gradient, correct?
Thank you very much for explanation! But how in Excel can we create a table that you shown at 1.38, 16.38+-0.3 with st.dev.?
Can explain the desirability functions with R code. I have 4 Factors and two responses
Thanks for the great tutorial! My question is: is it possible to add another level/layer? Eg to have: survived - class - age?
Barely understand your accent
As a native English speaker, it is perfectly. As is her explanation.
thank you for this video. It's very useful.
What accent is that ? excuse me
Smart way! It's much simpler than the method I learnt to compare down and up etc. Thank you!
each time I run this command multcompLetters4(Anova, Tukey) I receive this error msg : Error in `[.data.frame`(data, , fm[[1]]) : undefined columns selected, could you help me solving this problem please?
Thank you for the awesome videos. Can you just get me to understand how did you arrive to the axial points of 92.07, 77.93, 182.07 and 167.93? I have tried many times but couldn't get the same answer... Thank you in advance
Hello, when x = 1 there is a difference of 5°C or 5 minutes which means 1 = 5 so in order to calculate how much is 1,414 you just have to do 1,414 * 5 and you find 7,07. I hope this is clear.
I asked that question to Chat GPT
Thank you . great video.
I need to label y axis with "Regeneration percentage (%)" but in the code I can't write % because shows me an error. How can I put % in y axis? 🙏🏽
Thank you very much. Congratulations!!!❤
Certainly a great video!!!! However, I'm facing some problems with plot generation. At the end of line 39 you introduce 'y=quant', what is this? When I run the function for plot, it's generated an error 'quant object not found'. I would appreciate information about it. Thanks beforehand
How can I use fixed values as my data. let's say I already have the % values for each category?
This is amazing. Have been studying and working with experimental design for about 2 years. Never ever have I seen such comprehensive introduction as this video.
Great to hear! Thank you!!!
lovely teaching video! how to analysis soil fertility status?
EXCELLENT explanations !!!
why does she sounds like Sofia vergara ......nice explanation though.
very helpful!!!!
Rosane Your video is a lovely one and go ahead please on soil fertility status and different fertilizer rate like phosphorus and potassium etc.
dr. Rech your teaching is a lovely one!! please do on fertilizer rate of phosphorus and potassium by RCBD design
Thanks for the nice tutorial. how to reorder or rearrange the x variables mean feed types??
At last got some clarity from your video
hi, I'm working on laser-assisted turning and currently experimenting with five parameters: depth of cut, feed rate, spindle speed, laser power, and laser angle (the distance between the laser spot and cutting edge). Initially, I opted for a face-centered central composite design. However, I've noticed that when the laser power is set to 0 (indicating the laser is off), the laser angle becomes irrelevant. I'm uncertain about the appropriate experimental design in this situation. Could you please suggest a method or name for this scenario, considering the disregard for laser angle when laser power is zero? Thank you for your insights!
Hi Amir, you can send me an email using the contact on the channel details, so we can discuss your experiment.
hi, I'm working on laser-assisted turning and currently experimenting with five parameters: depth of cut, feed rate, spindle speed, laser power, and laser angle (the distance between the laser spot and cutting edge). Initially, I opted for a face-centered central composite design. However, I've noticed that when the laser power is set to 0 (indicating the laser is off), the laser angle becomes irrelevant. I'm uncertain about the appropriate experimental design in this situation. Could you please suggest a method or name for this scenario, considering the disregard for laser angle when laser power is zero? Thank you for your insights!
Wouldn't this be an interaction between variables? Sorry if that doesn't directly help, but just trying to clarify. Hope you find a good approach.
I'd go for grey box modeling. RSM is usually best when modeling the relationsships is very difficult. Your machine has some very dependend relationships, such as depth of cut, feed rate and spindle speed all influencing the energy input into the workpiece. Spend some time to build a grey box with relationships and unknown parameters, then do classic parameter identification.