@@venkat2132 I am glad It was helpful!

Yes, you can adapt it

I am using long table in twocolumn format, but it is not spanned to next page

# file aux_gdp_plot.R: library("readxl") # import Excel files into R library("curl") # connection interface library("tidyr") # data management library("dplyr") # data management library("ggplot2") # plot library("scales") # graphical tools # World Bank GDP ---- ## "data.worldbank.org/indicator/NY.GDP.MKTP.CD" ## link in the description url <- "api.worldbank.org/v2/en/indicator/NY.GDP.MKTP.CD?downloadformat=excel" destfile <- "NY_GDP_MKTP.xls" curl::curl_download(url, destfile) GDP <- read_excel(destfile, skip = 2) gdp <- GDP gdp <- gdp[, c(-2, -3, -4)] colnames(gdp)[1] <- "Country" # wide => long with pivot_longer() (tidyr) ---- gdp_l <- gdp %>% pivot_longer(!Country, # all columns but not country names_to = "year", values_to = "GDP") # file app.R #' In this video we write an app to plot countries' GDP by using World Bank data library("shiny") source("aux_gdp_plot.R", local = TRUE) # Define UI for application ui <- fluidPage( # Application title titlePanel("GDP per country, 1960 ~ , World Bank Data"), # first row divided in 2 parts fluidRow( column(6, selectInput("country", "Select Countries: ", choices = gdp$Country, multiple = TRUE, width = "100%")), column(6, selectInput("year", "Select year: ", choices = unique(gdp_l$year), multiple = TRUE, width = "100%"))), # second row with one part fluidRow(column(12, actionButton("p", "Plot", class = "btn-block"))), # third row with 2 parts fluidRow(column(8, plotOutput("distPlot", brush = "plot_brush")), column(4, tableOutput("data"))) ) # Define server logic required to draw a line plot with ggplot2 server <- function(input, output) { out <- eventReactive(input$p, { req(input$country, input$year) df <- subset(gdp_l, Country %in% input$country & year %in% input$year) }) output$distPlot <- renderPlot({ ggplot(out(), aes(x = year, y = GDP, group = Country, color = Country)) + geom_line(size = 1) + scale_y_continuous(labels = scales::dollar, breaks = pretty_breaks()) + scale_x_discrete(breaks = pretty_breaks()) + theme_classic() + theme(legend.position = "bottom", legend.title = element_blank()) }, res = 96) output$data <- renderTable({ brushedPoints(out(), input$plot_brush) }) } # Run the application shinyApp(ui = ui, server = server)

Thank you very much!

Hi, thank you! I understand that if you don't add the photos you get the expected result. If you add the images the location points disappear but the images are not loaded. But you don't get any error message. Is that correct? Is that possible that they are overlapping the location point but they are not shown? If so, double check the path to the images and their format. Please let me know if you solve it

@@labtimeRP This is my code: #Let's add icon photos: iconSet <- iconList( BRL_stadium = makeIcon("~/Yas-New-Directory-Aus/German_Photos/German_Stadiums/Olympiastadion (Berlin) .jpg ", 80,80), MUN_stadium = makeIcon("~/Yas-New-Directory-Aus/German_Photos/German_Stadiums/Allian Arena (Munich).jpg ", 80,80), DRT_stadium = makeIcon(makeIcon("~/Yas-New-Directory-Aus/German_Photos/German_Stadiums/Signal-Iduna Park (Dortmund) .jpg ", 80,80)), STU_stadium = makeIcon("~/Yas-New-Directory-Aus/German_Photos/German_Stadiums/Mercedes Benz Arena (Stuttgart).jpg ", 80,80), GEN_stadium = makeIcon("~/Yas-New-Directory-Aus/German_Photos/German_Stadiums/Veltins-Arena (Gelsenkirchen).jpg ", 80,80), FRF_stadium = makeIcon("~/Yas-New-Directory-Aus/German_Photos/German_Stadiums/Deutsche Bank Park (Frankfurt).jpg ", 80,80), HAM_stadium = makeIcon("~/Yas-New-Directory-Aus/German_Photos/German_Stadiums/Volksparkstadion (Hamburg).jpg ", 80,80), KAI_stadium = makeIcon("~/Yas-New-Directory-Aus/German_Photos/German_Stadiums/Fritz-Walter Stadion (Kaiserslautern).jpg ", 80,80), COL_stadium = makeIcon("~/Yas-New-Directory-Aus/German_Photos/German_Stadiums/Rhein-Energie Stadion (Cologne).jpg ", 80,80), HAN_stadium = makeIcon("~/Yas-New-Directory-Aus/German_Photos/German_Stadiums/HDI Arena (Hannover).jpg ", 80,80), LEI_stadium = makeIcon("~/Yas-New-Directory-Aus/German_Photos/German_Stadiums/Red Bull Arena (Leipzig).jpg ", 80,80), NUR_stadium = makeIcon("~/Yas-New-Directory-Aus/German_Photos/German_Stadiums/Max-Morlock Stadion (Nuremberg).jpg ", 80,80) ) leaflet()%>% addTiles()%>% addMarkers(lng = GER_stadium$lng, lat = GER_stadium$lat, icon = iconSet)%>% addProviderTiles(provider = "Stamen") It still won't show images! Not even location points.

Hi, I solved it. It was giving the same problem to me. I think it is due to an update. Use the following code: iconSet <- list( BRL_stadium = "germany_stadium/berlin_stadium.jpg", MUN_stadium = "germany_stadium/munich_stadium.jpg", DRT_stadium = "germany_stadium/dortmund_stadium.jpg", STU_stadium = "germany_stadium/stuttgart_stadium.jpg", GEN_statium = "germany_stadium/gelsenkirchen_stadium.jpg", FRF_stadium = "germany_stadium/frankfurt_stadium.jpg", HAM_stadium = "germany_stadium/hamburg_stadium.jpg", KAI_stadium = "germany_stadium/kaiserslautern_stadium.jpg", COL_stadium = "germany_stadium/cologne_stadium.jpg", HAN_stadium = "germany_stadium/hannover_stadium.jpg", LEI_stadium = "germany_stadium/leipzig_stadium.jpg", NUR_stadium = "germany_stadium/nuremberg_stadium.jpg" ) leaflet() %>% addTiles() %>% addMarkers(data = GER_stadium[, 3:4], icon = ~ icons( iconUrl = iconSet, iconWidth = 80, iconHeight = 80 )) Let me know if it works for you as well

@@labtimeRP Running the last part shows me this error: Error in guessLatLongCols(names(obj)) : Couldn't infer longitude/latitude columns

@@yasamanjafari9712 in my data frame the columns have lat and lng as column names

Hi, thank you. I do not have now the possibility to share it.

Hi, something like this? Use a recursive function to make R decide about the coordinates. Then pass them to the matrix and assign value 1

Hi, did you watch video number 6 in the playlist? Is that video not helpful for your problem?

Hi, can you tell me the error you got?

Hi, the approach is very naive. A while loop runs as long as the sum of the 1s in the matrix in less than the value we choose for n. n is also the end of a sequence that is stored in s. In the while loop R extracts randomly with replacement a value for the i index and a value for the j index from the s sequence. This makes sure that the selection is in the margin of the matrix. A 1 is assigned to [i,j] every time the loop runs. When the sum of 1s in the matrix is equal to n the loop stops. In the second part of the video, I will explain it again.

Hi, when you get "undefined columns selected" error means that you are trying to select columns that don't exist in the data frame. The error may be due to wrong position of the comma, wrong index number or column name, or the data frame created is wrong. I hope this helps

Hi, I pasted it below. I hope it is useful library("ggplot2") # to plot with ggplot library("sp") # coordinates library("raster") # geographic data analysis and modeling library("broom") # to make spatial obj into dataframe => to plot with ggplot # download geo data from GADM ---- jpn1 <- getData("GADM", country = "JPN", level = 1) # prefectures jpn2 <- getData("GADM", country = "JPN", level = 2) # town/city View(jpn1) head(jpn2) # plot with plot() plot(jpn1) plot(jpn2) # make data as data frame to plot with ggplot() ---- jpn1_df <- tidy(jpn1, region = "NAME_1") jpn2_df <- tidy(jpn2, region = "NAME_2") jpn2_df_try <- tidy(jpn2, region = "NAME_2", id = "GID_2") jpn2_df_try jpn2_df # generate some random values to plot ---- id <- unique(jpn2_df$id) id set.seed(123) rval <- sample(100:1000, length(id), replace = TRUE) df_city <- data.frame(id, rval) head(df_city) tail(df_city) df_city[which.max(df_city$rval), ] df_city[which.min(df_city$rval), ] # merge with jpn2_df by id ---- df <- merge(jpn2_df, df_city, by = "id") head(df) tail(df) # plot with ggplot() ---- ## plot prefectures ggplot() + geom_polygon(data = jpn1_df, aes(x = long, y = lat, group = group, fill = id)) + ggtitle("Japan Prefectures Map with ggplot2") + theme_void() + theme(legend.position = "none") ## plot cities ggplot() + geom_polygon(data = df, aes(x = long, y = lat, group = group, fill = id)) + ggtitle("Japan Metropolitan Map with ggplot2") + theme_void() + theme(legend.position = "none") ## overlap the two plots ---- ### plot random values at city level and overlap prefecture border lines ggplot() + geom_polygon(data = df, aes(x = long, y = lat, group = group, fill = rval)) + geom_polygon(data = jpn1_df, aes(x = long, y = lat, group = group), fill = "white", alpha = 0.1, color = "red") + ggtitle("Japan Cities & Prefectures Map with ggplot2") + theme_void() # Nakagawa plot ---- #' There are 4 Nakagawa cities #' in Fukuoka #' in Hokkaido #' in Nagano #' in Tochigi #' jpn2_df jpn2_df_try <- jpn2_df jpn2_df_try jpn2_df$newId <- ifelse(jpn2_df$group == "Nakagawa.1", "Hokkaido", ifelse(jpn2_df$group == "Nakagawa.2", "Tochigi", ifelse(jpn2_df$group == "Nakagawa.3", "Fukuoka", ifelse(jpn2_df$group == "Nakagawa.4", "Nagano", "NotNakagawa")))) jpn2_df jpn2_df$newId <- as.factor(jpn2_df$newId) ## plot Nakagawa ggplot() + geom_polygon(data = jpn2_df, aes(x = long, y = lat, group = group, fill = newId)) + ggtitle("Nakagawa cities in Japan") + theme_void() + coord_equal(xlim = c(130, 142), ylim = c(32, 44.5))

Thank you for appreciating my video 👍

Hi, if you refer to the code for the set up of the file you can find it in the description box of the second video in the playlist. I hope it helps

Hi, thank you for the comment. I would recommend that you place the python file in the folder where your tex file is located. The R file in the example is in the same folder. Usually, I organize files in folders in the same folder where the tex file is located. For an example, you may refer to the video about including figures in tex: kzbin.info/www/bejne/j4OZZoOOZsSnoM0 I hope it helps. I will discuss more about files organization in future video about how to write a thesis/book in LaTeX.

Thank you for your comment!

Hi, chordDiagram is a function from the circlize package. Make it sure that the package is installed and loaded

Hi, what kind of error do you get? Following the code library("shiny") library("ggplot2") library("quantmod") library("scales") library("dplyr") library("tidyr") library("tibble") library("TTR") # Define UI for application ui <- fluidPage( # Application title titlePanel("Stock Analysis"), # Sidebar sidebarLayout( sidebarPanel( textInput("ticker", "Ticker: "), dateInput("date", "Starting date: ", value = "2022-01-03"), selectInput("dateBreaks", "Date breaks: ", choices = c("1 day", "1 week", "1 month", "1 year")), selectInput("ma", "Moving average: ", choices = c("None", "SMA", "EMA", "WMA", "EVWMA", "VWAP")), numericInput("window", "window: ", value = 1), actionButton("button", "Compute", class = "btn-block") ), # Show plot and data mainPanel( fluidRow( column(12, plotOutput("candlePlot", brush = "plot_brush")) ), fluidRow( column(12, tableOutput("data")) ) ) ) ) # Define server logic server <- function(input, output, session) { stock <- eventReactive(input$button, { req(input$ticker, input$date, input$ma, input$window) getSymbols(input$ticker, from = input$date, auto.assign = FALSE) %>% na.locf() %>% as.data.frame() %>% `colnames<-`(c("open", "high", "low", "close", "volume", "adjusted")) %>% rownames_to_column(var = "date") %>% mutate(date = as.Date(date)) %>% mutate(greenRed=ifelse(open-close>0, "Red", "Green")) %>% mutate(movAvg = case_when( # new code to compute MA input$ma == "SMA" ~ SMA(close, input$window), input$ma == "EMA" ~ EMA(close, input$window), input$ma == "WMA" ~ WMA(close, input$window), input$ma == "EVWMA" ~ EVWMA(close, volume, input$window), input$ma == "VWAP" ~ VWAP(close, volume, input$window) )) } ) cp <- eventReactive(input$button, { req(input$ticker, input$date, input$dateBreaks) # Code for candlestick plot with ggplot2 from: # www.r-bloggers.com/2021/09/robservations-12-making-a-candlestick-plot-with-the-ggplot2-and-tidyquant-packages/ ggplot(stock()) + geom_segment(aes(x = date, xend = date, y = open, yend = close, colour = greenRed), size = 3) + geom_segment(aes(x = date, xend = date, y = high, yend = low, colour = greenRed)) + geom_col(aes(x = date, y = (((max(close) - min(close)) * ((volume - min(volume)) / (max(volume) - min(volume)))) + min(close))), # To constrain volume to be in the range of close: (((max(new) - min(new)) * ((x - min(old)) / (max(old) - min(old)))) + min(new))), fill = "red", alpha = 0.1) + theme_bw() + scale_color_manual(values = c("Forest Green","Red")) + labs(title = paste0(input$ticker, ": from ", stock()$date[1], " to ", stock()$date[nrow(stock())]), subtitle = paste0("Close price: $", round(stock()$close[nrow(stock())], 2), " ", "Close date: ", stock()$date[nrow(stock())])) + theme(legend.position ="none", axis.title.y = element_blank(), axis.title.x=element_blank(), axis.text.x = element_text(angle = 90, vjust = 0.5, hjust=1), plot.title= element_text(hjust=0.5)) + scale_y_continuous(labels = scales::label_dollar(), position = "right") + scale_x_date(date_breaks = input$dateBreaks, date_labels = "%Y-%m-%d") + coord_cartesian(ylim = c(min(stock()$close)*0.9, max(stock()$close))*1.05) + geom_line(aes(x = date, y = movAvg)) # new code to plot the MA }) output$candlePlot <- renderPlot({ cp() }, res = 96) output$data <- renderTable({ brushedPoints(stock(), input$plot_brush) }) } # Run the application shinyApp(ui = ui, server = server)

Hi, thank you for watching. Please note that the videos starting with "Write Functions in R" are about coding from scratch. This means that per_change() is a function that we write and it doesn't exist in R. PS I think you can find a function for percentage change in the TTR package.

Could not find per_change in TTR package, I am not clear how you write per_change in R. Could you explain further how a function that do not exist in R is written? Thanks.

Thank you for appreciating. If you like this kind of videos, next it's coming "battleship"

Thank you for your comment. Appreciated!

Hi, thank you for your feedback. Which result you can't get?

Hi, if I am not wrong it is an issue of shiny itself. However, I remembered I solved the issue in a later script. I think this series has 4 or 5 more scripts to go

hi, it will take a while to load the rest of the series. Basically, to solve that issue I added a new column in the data frame where I convert the date into a character. However, in the final script I completely remove the table output because I make the plot interactive with plotly so the data will be readable directly from the plot. I hope this helps

Thank you!

Hi, thank you for watching. Can you tell me if you get the same issue if you try to map a column of random values to town names? If you did not watch yet, I would recommend watching Maps with R: Tokyo map with ggplot2 | GADM geo data || 10 kzbin.info/www/bejne/oqPVnISipa1_eLM where I show how you can work with town data

@@labtimeRP Thank you for the advice. The town worked out fine. I was wondering if it was because I missed data when mapping my values to the name of cities.

Yes, I think that may be the issue. Working with town data is tricky. For example, if you try to map Tokyo you will not find it in the GADM data because Tokyo is reported by its special wards. Additionally, in the case of Japan special characters in the town name such as bar accent can create problems when mapping or merging.

Totally agree and tidying up the town names is really a torture... Thank you for the kind help. Is there any way to draw the broadline? I mean, the city border, so that the data can start from 0 and the place is colored white?

To trace a border line you can add color and size in the geom_polygon(). You may refer to the answer to one of the comments in this video Maps with R: choropleth map with ggplot2 | map of China, Japan and South Korea || 01 kzbin.info/www/bejne/gICaf2BqhZelaq8 For the city name, you can create a vector of ordered unique values and assign an id to each of them. Then you assign the same id to the unique entries in the other dataset. And then merge the two datasets by the id. Naturally I assume you have the same cities in the two datasets. It should not matter if they are written differently - i.e. some use special characters - when you sort them in alphabetical order. For Tokyo you may assign the same id to its wards. In this case when you merge all the Tokyo wards will be mapped to the Tokyo population. I hope this can be useful

I am glad you found it helpful 👍

Hi, thank you for watching. It seems to me that the KZbin algorithm is pushing the video in these days. However, please note that when the video was made most of the analysts I read about did not expect it would last so long and the word "conflict" was also used. However, I am not an expert of the region or the Russia-Ukraine relations. I updated the title accordingly.

Hi, thank you for watching. It seems to me that the KZbin algorithm is pushing the video in these days. However, please note that when the video was made most of the analysts I read about did not expect it would last so long and the word "conflict" was also used. However, I am not an expert of the region or the Russia-Ukraine relations. I updated the title accordingly.

Hi, thank you! If I understand correctly you want to add a border line between the countries, correct? If this is what you want you can add color and size. For example, in the choropleth map in the video replace this code: geom_polygon(data = df, aes(x = long, y = lat, group = group, fill = pop), colour = "red", size = 1) I hope this is helpful

@@labtimeRP Hi, thank you. Yes was exactly what i wanted to do.

@@MegaMaster764 perfect! 👍

Thank you for your comment! I am glad you found it interesting

Thank you for your comment. Very much appreciated!

Thank you for watching! I am glad you find it useful

Hi, thank you for watching. You can find the code in the description.

Sure, I will put in the description in the next days

Hi Adrien, thank you for watching. Perhaps the following videos answer your question: Maps with R: Tokyo map with ggplot2 | GADM geo data || 10 kzbin.info/www/bejne/oqPVnISipa1_eLM Maps with R: map of the Russia-Ukraine conflict with leaflet | GADM | video 2 || 13 kzbin.info/www/bejne/p52spXiljNOripI

@@labtimeRP thank you for the prompt answer. I will check that :).

Hi, thank you for watching. Which error do you get? Did you try also method 1?

Hi, I made a video to answer your question. I hope it is useful. kzbin.info/www/bejne/iXqkpq2od5iCibc