Читать книгу The Big R-Book - Philippe J. S. De Brouwer - Страница 4

1 Chapter 2Figure 2.1: A view on the steps in the scientific method for the data scient...Figure 2.2: The role of data-science in a company is to take data and turn i...

2 Chapter 3Figure 3.1: An example showing the histogram of data generated from the norm...

3 Chapter 4Figure 4.1: The plot-function will result in a bar-chart for a factor-object...Figure 4.2: The factor objects appear now in a logical order. Figure 4.3: The standard plot for a data frame in R shows each column printe...Figure 4.4: The histogram of the CAD. Figure 4.5: A scatter-plot of one variable with another. Figure 4.6: The histogram of the most recent values of the CAD only.

4 Chapter 7Figure 7.1: The sum of sine and cosine illustrated.Figure 7.2: A tibble plots itself like a data-frame. Figure 7.3: A linear model fit on generated data to illustrate the piping c...

5 Chapter 8Figure 8.1: A comparison between a set of random numbers drawn from the nor...Figure 8.2: The same plot for the returns of the SP500 index seems acceptab...Figure 8.3: A Q-Q plot is a good way to judge if a set of observations is n...Figure 8.4: The probability to get maximum x tails when flipping a fair coi...

6 Chapter 9Figure 9.1: The plot-function will generate a scatter-plot for a vector. Not...Figure 9.2: The plot-function will generate a scatter-plot of each column in...Figure 9.3: Some plot characters. Most other characters will just plot thems...Figure 9.4: A scatter-plot needs an x and a y variable. Figure 9.5: A line plot of the type b. Figure 9.6: A pie-chart in R. Figure 9.7: A standard bar-chart based on a vector. Figure 9.8: A bar-chart based on a matrix will produce stacked bars. Note ho...Figure 9.9: A boxplot where the total of each bar equals 100%. Note how the ...Figure 9.10: Boxplots show information about the central tendency (median) a...Figure 9.11: Violin plot as provided by the function vioplot from the packa...Figure 9.12: Violin plot as traced by geom_violin provided by the library ...Figure 9.13: Violin plot as traced by geom_violin provided by the library ...Figure 9.14: A histogram in R is produced by the hist() function. Figure 9.15: In this histogram, the breaks are changed, and the y-axes is no...Figure 9.16: Two line plots plotted by the function curve(). Figure 9.17: A colour mapping combined with a contour plot provides a nice i...Figure 9.18: Heatmap for the “mtcars” data. Figure 9.19: Heatmap for the “mtcars” data with all columns rescaled...Figure 9.20: The frequency of the ten most occurring words in this text. Not...Figure 9.21: A word-could for the text of this book. This single image gives...Figure 9.22: An illustration of six predefined colour schemes in R. This fig...Figure 9.23: A visualisation of all built in colours in R. Note that the num...Figure 9.24: Examples of discrete colour sets. The name of the colour-set is...

7 Chapter 10Figure 10.1: The standard plot for a time series object for the returns of t...Figure 10.2: The standard plot functionality of time series will keep the z-...Figure 10.3: Afirst plot to show the data beforewe start. Thiswill allow us ...Figure 10.4: A forecast based on moving average. Figure 10.5: A backtest for our forecast. Figure 10.6: Optimal moving average forecast. Figure 10.7: Forecasting with an exponentially smoothed moving average. Figure 10.8: Holt exponentially smoothed moving average. Figure 10.9: Using the stl-function to decompose data in a seasonal part and...Figure 10.10: TheHolt-Wintersmodel fits an exponential trend.Here we plot th...Figure 10.11: The Holt-Winters model applied to the temperatures in Nottingh...

8 Chapter 14Figure 14.1: The entity relationship (ER) diagram for our example, the libr...Figure 14.2: The database scheme. Figure 14.3: Different join types illustrated. Note that the Venn-diagrams ...Figure 14.4: The improved database scheme that allows multiple authors to c...

9 Chapter 15Figure 15.1: Histogram generated with data from the MySQL database.

10 Chapter 17Figure 17.1: Finally, we are able to make a plot of the tibble in a way that...Figure 17.2: The standard plot function on a factored object with some value...Figure 17.3: Maybe you would prefer to show this plot to the board meeting? ...Figure 17.4: A visualisation of how the age of customers impacted the satisf...

11 Chapter 18Figure 18.1: The visualization of missing data with the function md.pattern...

12 Chapter 19Figure 19.1: Two histograms of the same dataset. The histogram with less bin...Figure 19.2: A plot of the fabricated dataset with the spending ratio in fun...Figure 19.3: A simple aid to select binning borders is plotting a non-parame...Figure 19.4: The underlying relation between spending probability for femal...Figure 19.5: The dataset “as received from the customer service department”...Figure 19.6: The data does not reveal much patterns for any of the variable...

13 Chapter 20Figure 20.1: A visualization of the loadings of the principal components of ...Figure 20.2: The biplot of the datasetmtcars: all observation and dimension...Figure 20.3: Visual aids to select the optimal number of factors.

14 Chapter 21Figure 21.1: A scatter-plot generated by the line “plot(survey$Height, surve...Figure 21.2: A plot visualizing the linear regression model (the data in red...Figure 21.3: Using the function abline() and cleaning up the titles. Figure 21.4: The results of the non-linear regression with nls(). This plot...

15 Chapter 22Figure 22.1: The grey diamonds with red border are the data-points (not pass...Figure 22.2: The ROC curve of a logistic regression. Figure 22.3: The ROC curve plotted with ggplot2. Figure 22.4: A plot of the accuracy in function of the cut-off (threshold) l...Figure 22.5: The area under the curve (AUC) is the area A plus the area C. I...Figure 22.6: The KS as the maximum distance between the cumulative distribut...Figure 22.7: The KS as the maximum distance between the model and a pure ran...Figure 22.8: The cost functions compared different cost structures. In plot ...

16 Chapter 23Figure 23.1: An example of the decision tree on fake data a represented in t...Figure 23.2: Three alternatives for the impurity measure in the case of clas...Figure 23.3: The plot of the complexity parameter (cp) via the function plot...Figure 23.4: The decision tree, fitted by rpart. This figure helps to visual...Figure 23.5: The same tree as in Figure 23.4 but now pruned with a complexit...Figure 23.6: The decision tree represented by the function prp() from the pa...Figure 23.7: The plot of the complexity parameter (cp) via the function plot...Figure 23.8: rpart tree on mpg for the dataset mtcars. Figure 23.9: The same tree as in Figure 23.8 but now pruned with a complexit...Figure 23.10: ROC curve of the decision tree. Figure 23.11: The accuracy for the decision tree on the Titanic data. Figure 23.12: The plot of a randomForest object shows how the model improves...Figure 23.13: The importance of each variable in the random-forest model. Figure 23.14: Partial dependence on the variables (1 of 3). Figure 23.15: Partial dependence on the variables (2 of 3). Figure 23.16: Partial dependence on the variables (3 of 3). Figure 23.17: A logistic regression is actually a neural network with one ne...Figure 23.18: A simple neural net fitted to the dataset of mtcars, predictin...Figure 23.19: A visualisation of the ANN. Note that we left out the weights,...Figure 23.20: A visualisation of the performance of the ANN (left) compared ...Figure 23.21: Avisualisation of the performance of theANNcompared to the lin...Figure 23.22: A boxplot for the MSE of the cross validation for the ANN. Figure 23.23: The cars in the dataset mtcars with fuel consumption plotted i...Figure 23.24: The result of k-means clustering with three clusters on the we...Figure 23.25: The plot() function applied on a prcomp object visualises the ...Figure 23.26: The custom function biplot() project all data in the plane tha...Figure 23.27: A projection in the plane of the two major principal component...Figure 23.28: The projection of mtcars in the surface formed by the two firs...Figure 23.29: Two dimensional projections of the dependency structure of the...Figure 23.30: A three dimensional plot of the cars with on the z-axis the fi...Figure 23.31: plotly will produce a graph that is not only 3D but is interac...Figure 23.32: A plot with autoplot(), enhanced with ggrepel of the fuzzy clu...Figure 23.33: A hierarchical cluster for the dataset mtcars.

17 Chapter 24Figure 24.1: The results of the bootstrap exercise: a set of estimates for ...

18 Chapter 25Figure 25.1: A spacing grid for the predictions of t mpg. Figure 25.2: Bootstrapping the returns of the S&P500 index. Figure 25.3: The histograms of the different coefficients of the linear reg...Figure 25.4: The histogram of the RMSE for a Monte Carlo cross validation o...Figure 25.5: Histogramof the RMSE based on a 5-fold cross validation. The h...Figure 25.6: The life cycle of a model: a model is an integrated part of bu...

19 Chapter 26Figure 26.1: Demonstration of the barChart() function of the package quantm...Figure 26.2: Demonstration of the lineChart() function of the package quand...Figure 26.3: Demonstration of the candleChart() function of the package qua...Figure 26.4: Bollinger bands with the package quandmod. Figure 26.5: The evolution of the HSBC share for the last ten years....Figure 26.6: The Q-Q plot of our naive model to forecast the next opening p...

20 Chapter 27Figure 27.1: A visualization of the dominance relationship.Figure 27.2: The scores of different cities according to the WSM.Figure 27.3: The preference structure as found by the ELECTRE I method give...Figure 27.4: Another representation of Figure 27.3. It is clear that Krakow...Figure 27.5: The results of ELECTRE I with comparability index C2 and param...Figure 27.6: The results for ELECTRE Iwith comparability indexC2. The A → B...Figure 27.7: The preference structure as found by the ELECTRE II method giv...Figure 27.8: The results for ELECTRE I with comparability index C2.Figure 27.9: Examples of smooth transition schemes for preference functions...Figure 27.10: Examples of practically applicable preferences functions P(d)...Figure 27.11: The hierarchy between alternatives as found by PROMethEE I.Figure 27.12: The preference relations resulting from PROMethEE I. For exam...Figure 27.13: The result for PROMethEE I with different preference function...Figure 27.14: The results for PROMethEE I method with the custom preference...Figure 27.15: Promethee II can also be seen as using a richer preference st...Figure 27.16: The hierarchy between alternatives as found by PROMethEE II. ...Figure 27.17: PROMethEE II provides a full ranking. Here we show how much e...Figure 27.18: The variance explained by each principal component.Figure 27.19: A projection of the space of alternatives in the 2D‐plane for...Figure 27.20: A standard plot with autoplot() with labels coloured Figure 27.21: Autoplot with visualization of two clusters Figure 27.22: Clustering with elliptoid borders, labels of alternative, pro...

21 Chapter 28Figure 28.1: The elements of wealth creation in a company. The company acqu...Figure 28.2: KPIs of the Value Chain that can be used by a manager who want...

22 Chapter 30Figure 30.1: The Epachenikov kernel (left), for h = 1; and the Gaussian ker...Figure 30.2: As illustration on how the Epachenikov Kernel Estimation works...Figure 30.3: Some concepts illustrated on the example of a call option with...Figure 30.4: The intrinsic value of a long call illustratedwith its payoff ...Figure 30.5: The intrinsic value of a short call illustrated with its payof...Figure 30.6: The payoff and profit for a long put (left) and a short put (r...Figure 30.7: The price of a long call compared to its intrinsic value. The ...Figure 30.8: The price of a long put compared to its intrinsic value. Note ...Figure 30.9: Step 1 in the binomial model. Figure 30.10: The first 2 steps of the binomial model. Figure 30.11: The Cox–Ross–Rubinsteinmodel for the binomialmodel applied to...Figure 30.12: he Cox–Ross–Rubinsteinmodel for the binomialmodel applied to...Figure 30.13: The value of a call option depends on many variables. Some ar...Figure 30.14: The value of a put option depends on many variables. Some are...Figure 30.15: An illustration of how the delta of a call and put compare in...Figure 30.16: Linear option strategies illustrated. The red line is the int...Figure 30.17: Linear option strategies illustrated. Part 2 (basic composite...Figure 30.18: Linear option strategies illustrated. Part 3 (some more compl...Figure 30.19: A covered call is a short call where the losses are protected...Figure 30.20: A married put is a put option combined with the underlying as...Figure 30.21: A collar is a structure that protects us from strong downward...

23 Chapter 31Figure 31.1: A basic and simple scatter-plot generated with ggplot2. Figure 31.2: The same plot as in previous figure, but now enhanced with Loe...Figure 31.3: The same plot as in previous Figure, but now enhanced with dif...Figure 31.4: A facet plot will create sub-plots per discrete value of one o...Figure 31.5: The standard functionality for scatterplots is not optimal for...Figure 31.6: The contour plot is able to show where the density of points i...Figure 31.7: Adding a Loess estimate is a good idea to visualize the genera...Figure 31.8: This plot shows a facet plot of a contour plot with customised...

24 Chapter 32Figure 32.1: Selecting File → New File → R Markdown… in RStudio will ope...

25 Chapter 33Figure 33.1: The LATEX article looks like this. Note that this is a cropped ...

26 Chapter 36Figure 36.1: The output of one of the examples supplied by the Shiny packag...Figure 36.2: The web-page produced by the code fragment above. Figure 36.3: A map created by leaflet based on the famous OpenStreetMap m...Figure 36.4: A useful tool to explore new data and/or get an intuitive unde...Figure 36.5: An interactive app with ggvis. This particular example uses t...Figure 36.6: The evolution of the gender-diversity-index in function of one...Figure 36.7: Creating a flexdashboard from the template provides a useful b...Figure 36.8: The welcome page of the dashboard provides the overview (menu)...Figure 36.9: The page “Gender” of the dashboard provides multiple views of ...Figure 36.10: A simple dashboard with shinydashboard. Figure 36.11: Another take on the diversity dashboard, rendered with the he...

27 Chapter 37Figure 37.1: The runtimes for matrix multiplication compared on the CPU ver...Figure 37.2: The runtimes for matrix multiplication for storage/calculation...

28 Chapter 39Figure 39.1: The status of Spark can be controlled via a regular web-browse...Figure 39.2: Each user can check information about his own Spark connection...

29 Chapter 40Figure 40.1: The package microbenchmark also defines a suible visualisatio...Figure 40.2: This figure shows that working with vectors is fastest (f2())....Figure 40.3: Different ways to calculate . The definition of the functions ...Figure 40.4: Autoplot generates nice violin plots to compare the speed of t...Figure 40.5: A violin plot visualizing the advantage of using C++ code in R...Figure 40.6: A violin plot visualizing the advantage of using C++ code in R...Figure 40.7: A visualization of the profiling of our functions. Figure 40.8: A flame-plot produced by the function flameGraph() from the p...Figure 40.9: A Callee Tee Map produced by the function calleeTreeMap() fro...

30 Appendix DFigure D.1: A visual aid to select binning borders is plotting a non-parametr...