APPLE INC: STOCK PRICE PREDICTION
TARANG JAIN
28 February 2018
CAPSTONE PROJECT ON STOCK PRICE PREDICTION OF APPLE INC. BASED ON HISTORICAL STOCK MARKET DATA FROM 1980-2018 LISTED ON NASDAQ
Made under the guidance of Prof. Sameer Mathur,IIM Lucknow as a part of the internship on Data Analytics with Managerial Applications.
1.INTRODUCTION
Predicting the Stock Market has been the bane and goal of investors since its existence. Everyday billions of dollars are traded on the exchange, and behind each dollar is an investor hoping to profit in one way or another. Entire companies rise and fall daily based on the behaviour of the market. Should an investor be able to accurately predict market movements, it offers a tantalizing promises of wealth and influence.
The efficient-market hypothesis suggests that stock prices reflect all currently available information and any price changes that are not based on newly revealed information thus are inherently unpredictable.A hypothesis which can near about predict the closing price of a stock on a particular day can become very handy for successful trading.
2.OVERVIEW OF STUDY
Apple Inc. is an American multinational technology company headquartered in Cupertino, California that designs, develops, and sells consumer electronics, computer software, and online services.
Apple was founded by Steve Jobs, Steve Wozniak, and Ronald Wayne in April 1976 to develop and sell Wozniak’s Apple I personal computer. It was incorporated as Apple Computer, Inc. in January 1977, and sales of its computers, including the Apple II, saw significant momentum and revenue growth for the company.
Apple went public(IPO) in 1980 to instant financial success. Over the next few years, Apple shipped new computers featuring innovative graphical user interfaces, and Apple’s marketing commercials for its products received widespread critical acclaim.
Apple is the world’s largest information technology company by revenue and the world’s second-largest mobile phone manufacturer. In February 2015, Apple became the first U.S. company to be valued at over US$700 billion. Apple’s worldwide annual revenue totaled $229 billion for the 2017 fiscal year.
One of the most popular and fastest growing stock today is that of google.Thus in tis project I am trying to predict the future stock pricing of google by hypothesing a model based on the past stock data.
3.DESCRIPTION OF DATASET
The dataset has been sourced from ana online website. It is a comprehensive collection of daily records of stocks belonging to Apple INC listed on NASDAQ from 12-12-1980 to 27-02-2018.The meaning of the columns has been explained below:
1.Date
The particular date for which the stock details are listed.
2.Open
The opening price of stock on that date.
3.High
The highest price to which the stock rose on that date.
4.Low
The lowest price to which the stock went down on that date.
5.Close
The closing price of stock on that date.
6.Volume
The amount of stock traded on that date.
6.Dividend
A dividend is defined as a payment made by a corporation to its shareholders. Usually these payouts are made in cash (called “cash dividends”), but sometimes companies will also distribute stock dividends, whereby additional stock shares are distributed to shareholders. Stock dividends are also known as stock splits.
6.Split
When a company declares a stock split, the number of shares of that company increases, but the market cap remains the same. Existing shares split, but the underlying value remains the same. As the number of shares increases, price per share goes down. The split ratio is given in the dataset
7.Adjusted Open
An adjusted opening price is a stock’s opening price on any given day of trading that has been amended to include any distributions and corporate actions that occurred at any time prior to the last day’s close.
8.Adjusted High
The highest price to which the stock rose on that day adjusted for splits.
9.Adjusted Low
The lowest price to which the stock went down on that day adjusted for splits.
10.Adjusted Close
An adjusted closing price is a stock’s closing price on any given day of trading that has been amended to include any distributions and corporate actions that occurred at any time prior to the next day’s open
11.Adjusted Volume
When adjusting historical data for stock-splits, StockFetcher only adjusts the price, not volume data. Adjusting volume data creates “spikes” which can skew measures and create “false positives” in stock screens; when, in actuality, the volume “pattern” may have been continuous. Same as the volume.
4.Model(Hypothesis)
Based on the t- tests and chi square tests performed it was observed that the closing price of the stock depeneded on:-
- Volume
- Adjusted High
- Adjusted Low
- Adjusted Open
- Adjusted volume
5.Result
After running a linear regression on the hypotehsis stated it was observed that the hypothesis actually gained emeprical support from the model. And a linear regression model was formulated based on the hypothesis which can predict the closing price of stock using past data. The p-value was found to be <0.05 for the independent variables tested.
Further, the Multiple R-squared= 0.4657, Adjusted R-squared= 0.4654
6.Conclusion
Thus it can be concluded that the pricing of the stock is depenedent on the adjusted high,adjusted low,adjusted Open, volume and adjusted volume of the stock.
Thus the linear regression model can predict the price of Apple INC’s stock with the help of the historical stock data.
7.R Code and analysis
stock <- read.csv(paste("apple dataset.csv", sep=""))
summary(stock)## Date Open High Low
## 1980-12-12: 1 Min. : 11.12 Min. : 11.12 Min. : 11.00
## 1980-12-15: 1 1st Qu.: 27.00 1st Qu.: 27.50 1st Qu.: 26.50
## 1980-12-16: 1 Median : 43.75 Median : 44.50 Median : 43.00
## 1980-12-17: 1 Mean :101.09 Mean :102.32 Mean : 99.75
## 1980-12-18: 1 3rd Qu.:109.70 3rd Qu.:111.17 3rd Qu.:108.37
## 1980-12-19: 1 Max. :702.41 Max. :705.07 Max. :699.57
## (Other) :9376
## Close Volume Dividend Split
## Min. : 11.00 Min. : 4471 Min. :0.000000 Min. :1.000
## 1st Qu.: 27.00 1st Qu.: 1230750 1st Qu.:0.000000 1st Qu.:1.000
## Median : 43.75 Median : 3759200 Median :0.000000 Median :1.000
## Mean :101.05 Mean : 11962851 Mean :0.003774 Mean :1.001
## 3rd Qu.:109.78 3rd Qu.: 17908300 3rd Qu.:0.000000 3rd Qu.:1.000
## Max. :702.10 Max. :189560600 Max. :3.290000 Max. :7.000
##
## Adj_Open Adj_High Adj_Low
## Min. : 0.1623 Min. : 0.1623 Min. : 0.1605
## 1st Qu.: 0.9157 1st Qu.: 0.9329 1st Qu.: 0.8966
## Median : 1.4256 Median : 1.4539 Median : 1.3969
## Mean : 21.1167 Mean : 21.3165 Mean : 20.9023
## 3rd Qu.: 19.6046 3rd Qu.: 19.9391 3rd Qu.: 19.2787
## Max. :179.1000 Max. :180.4800 Max. :178.1600
##
## Adj_Close Adj_Volume
## Min. : 0.1605 Min. :2.504e+05
## 1st Qu.: 0.9151 1st Qu.:3.475e+07
## Median : 1.4248 Median :6.080e+07
## Mean : 21.1149 Mean :8.873e+07
## 3rd Qu.: 19.6084 3rd Qu.:1.111e+08
## Max. :178.9700 Max. :1.855e+09
## View(stock)
attach(stock)library(psych)
describe(stock)## vars n mean sd median trimmed
## Date* 1 9382 4691.50 2708.49 4691.50 4691.50
## Open 2 9382 101.09 135.26 43.75 65.94
## High 3 9382 102.32 136.38 44.50 66.93
## Low 4 9382 99.75 133.92 43.00 64.90
## Close 5 9382 101.05 135.18 43.75 65.93
## Volume 6 9382 11962850.54 16636897.77 3759200.00 8522664.50
## Dividend 7 9382 0.00 0.09 0.00 0.00
## Split 8 9382 1.00 0.06 1.00 1.00
## Adj_Open 9 9382 21.12 38.44 1.43 11.41
## Adj_High 10 9382 21.32 38.75 1.45 11.54
## Adj_Low 11 9382 20.90 38.12 1.40 11.26
## Adj_Close 12 9382 21.11 38.45 1.42 11.40
## Adj_Volume 13 9382 88725799.58 87086110.15 60796366.50 72971312.39
## mad min max range skew kurtosis
## Date* 3477.44 1.00 9.382000e+03 9.381000e+03 0.00 -1.20
## Open 32.81 11.12 7.024100e+02 6.912900e+02 2.42 5.23
## High 33.27 11.12 7.050700e+02 6.939500e+02 2.42 5.22
## Low 32.25 11.00 6.995700e+02 6.885700e+02 2.42 5.24
## Close 32.81 11.00 7.021000e+02 6.911000e+02 2.42 5.23
## Volume 4731940.29 4471.00 1.895606e+08 1.895561e+08 2.34 8.13
## Dividend 0.00 0.00 3.290000e+00 3.290000e+00 31.44 1040.79
## Split 0.00 1.00 7.000000e+00 6.000000e+00 87.07 8006.29
## Adj_Open 1.47 0.16 1.791000e+02 1.789400e+02 2.07 3.45
## Adj_High 1.50 0.16 1.804800e+02 1.803200e+02 2.07 3.44
## Adj_Low 1.44 0.16 1.781600e+02 1.780000e+02 2.08 3.48
## Adj_Close 1.47 0.16 1.789700e+02 1.788100e+02 2.08 3.46
## Adj_Volume 47243592.37 250376.00 1.855410e+09 1.855160e+09 3.42 28.71
## se
## Date* 27.96
## Open 1.40
## High 1.41
## Low 1.38
## Close 1.40
## Volume 171761.03
## Dividend 0.00
## Split 0.00
## Adj_Open 0.40
## Adj_High 0.40
## Adj_Low 0.39
## Adj_Close 0.40
## Adj_Volume 899085.88mytable <- with(stock, table(Dividend))
mytable## Dividend
## 0 0.08 0.1 0.11 0.12 0.47 0.52 0.57 0.63 2.65 3.05 3.29
## 9326 4 4 4 21 3 4 4 4 3 4 1mytable1 <- with(stock, table(Split))
mytable1## Split
## 1 2 7
## 9378 3 1attach(stock)## The following objects are masked from stock (pos = 4):
##
## Adj_Close, Adj_High, Adj_Low, Adj_Open, Adj_Volume, Close,
## Date, Dividend, High, Low, Open, Split, Volumelibrary(lattice)
histogram(~Open) 
histogram(~Close) 
boxplot(Open)
boxplot(Close)
boxplot(High)
boxplot(Low) 
histogram(~Volume)
plot(~Close + Date , main = "Closing price of stock with day number" , pch =1)
library(car)##
## Attaching package: 'car'## The following object is masked from 'package:psych':
##
## logitscatterplotMatrix(formula= ~ Close + Open + High + Low + Volume,cex=0.8)
cor(stock[, c(2:6 , 9:13)])## Open High Low Close Volume Adj_Open
## Open 1.0000000 0.9999281 0.9999002 0.9998257 0.2882580 0.53307853
## High 0.9999281 1.0000000 0.9998790 0.9999201 0.2895081 0.53210141
## Low 0.9999002 0.9998790 1.0000000 0.9999135 0.2860077 0.53424426
## Close 0.9998257 0.9999201 0.9999135 1.0000000 0.2878256 0.53316338
## Volume 0.2882580 0.2895081 0.2860077 0.2878256 1.0000000 0.61196816
## Adj_Open 0.5330785 0.5321014 0.5342443 0.5331634 0.6119682 1.00000000
## Adj_High 0.5334196 0.5325140 0.5346118 0.5335876 0.6139130 0.99995700
## Adj_Low 0.5320958 0.5311451 0.5333616 0.5322755 0.6090673 0.99994498
## Adj_Close 0.5326582 0.5317641 0.5339189 0.5329171 0.6113204 0.99990481
## Adj_Volume 0.1916405 0.1944722 0.1874925 0.1911254 0.5403455 -0.05206478
## Adj_High Adj_Low Adj_Close Adj_Volume
## Open 0.53341960 0.53209578 0.53265821 0.19164055
## High 0.53251400 0.53114505 0.53176410 0.19447224
## Low 0.53461177 0.53336156 0.53391886 0.18749247
## Close 0.53358757 0.53227548 0.53291712 0.19112537
## Volume 0.61391304 0.60906726 0.61132042 0.54034546
## Adj_Open 0.99995700 0.99994498 0.99990481 -0.05206478
## Adj_High 1.00000000 0.99993011 0.99995486 -0.05071799
## Adj_Low 0.99993011 1.00000000 0.99995482 -0.05426231
## Adj_Close 0.99995486 0.99995482 1.00000000 -0.05249642
## Adj_Volume -0.05071799 -0.05426231 -0.05249642 1.00000000library(corrgram)
corrgram(stock[, c(2:6 , 9:13)] , order = T, text.panel=panel.txt,lower.panel = panel.shade,upper.panel = panel.pie, main="Corrgram of all variables")
t.test(Open, Close)##
## Welch Two Sample t-test
##
## data: Open and Close
## t = 0.018981, df = 18762, p-value = 0.9849
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## -3.832215 3.907160
## sample estimates:
## mean of x mean of y
## 101.0878 101.0503t.test(Low , Close)##
## Welch Two Sample t-test
##
## data: Low and Close
## t = -0.66079, df = 18760, p-value = 0.5088
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## -5.148646 2.552448
## sample estimates:
## mean of x mean of y
## 99.75224 101.05034t.test(High , Close)##
## Welch Two Sample t-test
##
## data: High and Close
## t = 0.6408, df = 18761, p-value = 0.5217
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## -2.615459 5.156211
## sample estimates:
## mean of x mean of y
## 102.3207 101.0503t.test(Volume , Close)##
## Welch Two Sample t-test
##
## data: Volume and Close
## t = 69.648, df = 9381, p-value < 2.2e-16
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## 11626061 12299438
## sample estimates:
## mean of x mean of y
## 1.196285e+07 1.010503e+02t.test(Adj_High , Close)##
## Welch Two Sample t-test
##
## data: Adj_High and Close
## t = -54.921, df = 10912, p-value < 2.2e-16
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## -82.57958 -76.88808
## sample estimates:
## mean of x mean of y
## 21.31651 101.05034t.test(Adj_Low, Close)##
## Welch Two Sample t-test
##
## data: Adj_Low and Close
## t = -55.274, df = 10864, p-value < 2.2e-16
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## -82.99038 -77.30578
## sample estimates:
## mean of x mean of y
## 20.90226 101.05034t.test(Adj_Open , Close)##
## Welch Two Sample t-test
##
## data: Adj_Open and Close
## t = -55.092, df = 10888, p-value < 2.2e-16
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## -82.77773 -77.08962
## sample estimates:
## mean of x mean of y
## 21.11667 101.05034t.test(Adj_Volume, Close)##
## Welch Two Sample t-test
##
## data: Adj_Volume and Close
## t = 98.684, df = 9381, p-value < 2.2e-16
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## 86963295 90488102
## sample estimates:
## mean of x mean of y
## 8.872580e+07 1.010503e+02FROM THE T-TESTS PERFORMED ABOVE IT IS INFERRED THAT THE PRICING OF THE STOCK PRIMARILY DEPENDS ON ADJUSTED HIGH, ADJUSTED LOW, ADJUSTED OPEN, VOLUME, ADJUSTED VOLUME.
model <- lm(formula = Close ~ Volume + Adj_High + Adj_Open + Adj_Low + Adj_Volume)
summary(model)##
## Call:
## lm(formula = Close ~ Volume + Adj_High + Adj_Open + Adj_Low +
## Adj_Volume)
##
## Residuals:
## Min 1Q Median 3Q Max
## -859.61 -34.41 -10.76 11.93 429.87
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 1.357e+01 1.634e+00 8.307 <2e-16 ***
## Volume -6.064e-06 1.264e-07 -47.963 <2e-16 ***
## Adj_High 7.037e+01 3.274e+00 21.493 <2e-16 ***
## Adj_Open 6.232e+00 3.371e+00 1.849 0.0645 .
## Adj_Low -7.418e+01 3.017e+00 -24.590 <2e-16 ***
## Adj_Volume 8.915e-07 1.716e-08 51.944 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 98.84 on 9376 degrees of freedom
## Multiple R-squared: 0.4657, Adjusted R-squared: 0.4654
## F-statistic: 1634 on 5 and 9376 DF, p-value: < 2.2e-16plot(model)
