chapter 4

Author: tomi77

.gitignore

@@ -15,5 +15,8 @@ cabal.sandbox.config
 *.aux
 *.hp
 
-/earthquakes.sql
-.idea
+/*.sql
+/*.png
+/*.dat
+.idea
+.directory

LearningHaskellDataAnalysis.cabal

@@ -63,9 +63,9 @@ cabal-version:       >=1.10
 --  build-depends:       base >=4.8 && <4.9, Cabal >= 1.10, csv
 
 library
-  build-depends:       base >=4.8 && <4.9, csv, HDBC, sqlite, HDBC-sqlite3, regex-posix
+  build-depends:       base >=4.8 && <4.9, csv, HDBC, sqlite, HDBC-sqlite3, regex-posix, easyplot
   hs-source-dirs:      src
-  exposed-modules:     LearningHaskellDataAnalysis01, LearningHaskellDataAnalysis02, LearningHaskellDataAnalysis03
+  exposed-modules:     LearningHaskellDataAnalysis01, LearningHaskellDataAnalysis02, LearningHaskellDataAnalysis03, LearningHaskellDataAnalysis04
   exposed:             True
 
 executable median

README.md

@@ -91,3 +91,52 @@ better version
 fixed data
 
     identifyInCSVFileFromColumn (\ x -> not (x =~ "^[1-9][0-9]?/[1-9][0-9]?/[12][0-9][0-9][0-9]$")) "poordataFixed.csv" "Number" "Birthday"
+
+# Chapter 4
+
+    convertCSVFileToSQL "aapl.csv" "aapl.sql" "aapl" ["date STRING", "open REAL", "high REAL", "low REAL", "close REAL", "volume REAL", "adjclose REAL"]
+
+## pullStockClosingPrices
+
+    aapl <- pullStockClosingPrices "aapl.sql" "aapl"
+    plot (PNG "aapl.png") $ Data2D [Title "AAPL"] [] $ aapl
+    plot (PNG "aapl_line.png") $ Data2D [Title "AAPL", Style Lines] [] $ aapl
+    plot (PNG "aapl_oneyear.png") $ Data2D [Title "AAPL", Style Lines] [] $ take 252 aapl
+
+## applyPercentChangeToData
+
+    aapl <- pullStockClosingPrices "aapl.sql" "aapl"
+    let aapl252 = take 252 aapl
+    let aapl252pc = applyPercentChangeToData aapl252
+
+google
+
+    convertCSVFileToSQL "googl.csv" "googl.sql" "googl" ["date STRING", "open REAL", "high REAL", "low REAL", "close REAL", "volume REAL", "adjclose REAL"]
+    googl <- pullStockClosingPrices "googl.sql" "googl"
+    let googl252 = take 252 googl
+    let googl252pc = applyPercentChangeToData googl252
+
+microsoft
+
+    convertCSVFileToSQL "msft.csv" "msft.sql" "msft" ["date STRING", "open REAL", "high REAL", "low REAL", "close REAL", "volume REAL", "adjclose REAL"]
+    msft <- pullStockClosingPrices "msft.sql" "msft"
+    let msft252 = take 252 msft
+    let msft252pc = applyPercentChangeToData msft252
+
+all data
+
+    plot (PNG "aapl_googl_msft_pc.png") [Data2D [Title "AAPL - One Year, % Change", Style Lines, Color Red] [] aapl252pc, Data2D [Title "GOOGL - One Year, % Change", Style Lines, Color Blue] [] googl252pc, Data2D [Title "MSFT - One Year, % Change", Style Lines, Color Green] [] msft252pc]
+
+## applyMovingAverageToData
+
+    aapl <- pullStockClosingPrices "aapl.sql" "aapl"
+    let aapl252 = take 252 aapl
+    let aapl252pc = applyPercentChangeToData aapl252
+    let aapl252ma20 = applyMovingAverageToData aapl252pc 20
+    plot (PNG "aapl_20dayma.png") [Data2D [Title "AAPL - One Year, % Change", Style Lines, Color Red] [] aapl252pc, Data2D [Title "AAPL 20-Day MA", Style Lines, Color Black] [] aapl252ma20]
+
+earthquakes
+
+    convertCSVFileToSQL "all_month.csv" "earthquakes.sql" "oneMonth" ["time TEXT", "latitude REAL", "longitude REAL", "depth REAL", "mag REAL", "magType TEXT", "nst INTEGER", "gap REAL", "dmin REAL", "rms REAL", "net REAL", "id TEXT", "updated TEXT", "place TEXT", "type TEXT"]
+    coords <- pullLatitudeLongitude "earthquakes.sql" "oneMonth"
+    plot (PNG "earthquakes.png") [Data2D [Title "Earthquakes", Color Red, Style Dots] [] coords]

aapl.csv

@@ -0,0 +1,54 @@
+module LearningHaskellDataAnalysis04 where
+import Data.List
+import Database.HDBC.Sqlite3
+import Database.HDBC
+import Graphics.EasyPlot
+import LearningHaskellDataAnalysis02
+
+readIntegerColumn :: [[SqlValue]] -> Integer -> [Integer]
+readIntegerColumn sqlResult index = map (\ row -> fromSql $ genericIndex row index :: Integer) sqlResult
+
+readDoubleColumn :: [[SqlValue]] -> Integer -> [Double]
+readDoubleColumn sqlResult index = map (\ row -> fromSql $ genericIndex row index :: Double) sqlResult
+
+readStringColumn :: [[SqlValue]] -> Integer -> [String]
+readStringColumn sqlResult index = map (\ row -> fromSql $ genericIndex row index :: String) sqlResult
+
+queryDatabase :: FilePath -> String -> IO [[SqlValue]]
+queryDatabase databaseFile sqlQuery = do
+    conn <- connectSqlite3 databaseFile
+    result <- quickQuery' conn sqlQuery []
+    disconnect conn
+    return  result
+
+pullStockClosingPrices :: String -> String -> IO [(Double, Double)]
+pullStockClosingPrices databaseFile database = do
+    sqlResult <- queryDatabase databaseFile ("SELECT rowid, adjclose FROM " ++ database)
+    return $ zip (reverse $ readDoubleColumn sqlResult 0) (readDoubleColumn sqlResult 1)
+
+percentChange :: Double -> Double -> Double
+percentChange value first = 100.0 * (value - first) / first
+
+applyPercentChangeToData :: [(Double, Double)] -> [(Double, Double)]
+applyPercentChangeToData dataset = zip indices scaledData
+  where
+    (_, first) = last dataset
+    indices = reverse [1.0..(genericLength dataset)]
+    scaledData = map (\ (_, value) -> percentChange value first) dataset
+
+movingAverage :: [Double] -> Integer -> [Double]
+movingAverage values window =
+    if window >= genericLength values
+    then [average values]
+    else average (genericTake window values):(movingAverage (tail values) window)
+
+applyMovingAverageToData :: [(Double, Double)] -> Integer -> [(Double, Double)]
+applyMovingAverageToData dataset window =
+    zip [fromIntegral window..] $ movingAverage (map snd (reverse dataset)) window
+
+pullLatitudeLongitude :: String -> String -> IO [(Double, Double)]
+pullLatitudeLongitude databaseFile database = do
+  sqlResult <- queryDatabase
+    databaseFile
+    ("SELECT latitude, longitude FROM " ++ database)
+  return $ zip (readDoubleColumn sqlResult 1) (readDoubleColumn sqlResult 0)