Kordyuk_Lab_6_Report

library(tidyverse)
library(tsibble)
library(ggplot2)
library(lubridate)
library(readxl)
library(gridExtra)
library(dplyr)
library(patchwork)
library(zoo)
library(fable)
library(feasts)
library(fabletools)
library(urca)

tesla <- as_tsibble(read_csv("TSLA.csv"), index = Date)%>%
  glimpse()

Rows: 965
Columns: 7
$ Date        <date> 2019-01-02, 2019-01-03, 2019-01-04, 2019-01-07, 2019-01-0…
$ Open        <dbl> 20.40667, 20.46667, 20.40000, 21.44800, 22.79733, 22.36667…
$ High        <dbl> 21.00867, 20.62667, 21.20000, 22.44933, 22.93400, 22.90000…
$ Low         <dbl> 19.92000, 19.82533, 20.18200, 21.18333, 21.80133, 22.09800…
$ Close       <dbl> 20.67467, 20.02400, 21.17933, 22.33067, 22.35667, 22.56867…
$ `Adj Close` <dbl> 20.67467, 20.02400, 21.17933, 22.33067, 22.35667, 22.56867…
$ Volume      <dbl> 174879000, 104478000, 110911500, 113268000, 105127500, 814…

Заповнюємо пропуски

tesla = tesla %>%
  fill_gaps() %>% 
  fill(Open, .direction = "down") %>% 
  fill(High, .direction = "down") %>%
  fill(Low, .direction = "down") %>%
  fill(Close, .direction = "down") %>%
  fill(`Adj Close`, .direction = "down") %>%
  fill(Volume, .direction = "down")

tesla %>%
  filter(year(Date) == 2020) %>%
  autoplot(High) +
  labs(y = "Найвища щоденна ціна акції у 2020 році", x = "День")

tesla %>%
  filter(year(Date) == 2020) %>%
  autoplot(difference(High)) +
  labs(y = "Найвища щоденна ціна акції у 2020 році", x = "День")

tesla %>%
  ACF(High) %>% autoplot()

tesla %>%
  ACF(difference(High)) %>% autoplot()

tesla %>%
  autoplot(log(Volume))

tesla %>%
  autoplot(log(Volume) %>%
             difference(12))

KPSS ТЕСТ

tesla %>%
  features(High, unitroot_kpss)

# A tibble: 1 × 2
  kpss_stat kpss_pvalue
      <dbl>       <dbl>
1      15.8        0.01

tesla %>%
  features(High, unitroot_ndiffs)

# A tibble: 1 × 1
  ndiffs
   <int>
1      1

АВТОМАТИЧНИЙ ВИБІР ПОРЯДКУ РІЗНИЦЬ

tesla %>% mutate(log_sales = log(High)) %>%
  features(log_sales, list(unitroot_nsdiffs, feat_stl))

# A tibble: 1 × 10
  nsdiffs trend_stren…¹ seaso…² seaso…³ seaso…⁴ spikin…⁵ linea…⁶ curva…⁷ stl_e…⁸
    <int>         <dbl>   <dbl>   <dbl>   <dbl>    <dbl>   <dbl>   <dbl>   <dbl>
1       0          1.00   0.192       0       5 1.24e-12    40.1   -11.1   0.389
# … with 1 more variable: stl_e_acf10 <dbl>, and abbreviated variable names
#   ¹​trend_strength, ²​seasonal_strength_week, ³​seasonal_peak_week,
#   ⁴​seasonal_trough_week, ⁵​spikiness, ⁶​linearity, ⁷​curvature, ⁸​stl_e_acf1

ARIMA

tesla %>% filter(year(Date) == 2020) %>%
  model(ARIMA(Volume)) %>% 
  report()

Series: Volume 
Model: ARIMA(2,1,1)(2,0,1)[7] 

Coefficients:
         ar1      ar2      ma1    sar1    sar2     sma1
      0.7866  -0.1316  -0.9741  0.5099  0.0930  -0.5406
s.e.  0.0547   0.0537   0.0199  0.3445  0.0591   0.3431

sigma^2 estimated as 5.855e+15:  log likelihood=-7141.55
AIC=14297.11   AICc=14297.42   BIC=14324.41

tesla %>%
  ACF(High) %>%
  autoplot()

tesla %>%
  PACF(Volume) %>%
  autoplot()