macro.Rmd

---
title: "macro"
output:
  html_document: default
  pdf_document: default
date: '2023-09-19'
---

```{r setup, include=FALSE}
knitr::opts_chunk$set(echo = TRUE)

library(knitr)
 library(flipTime)
library(lubridate)
library(purrr)
library(sweep)
  library(dplyr)
  library(tidyr)
  library(tibbletime)
  library(timetk)
  library(tidyverse)
library(tidyquant)
library(readxl)
library(xts)
library(PerformanceAnalytics)
library(ggplot2)
library(ggfortify)
library(ggforce)
library(rvest)
library(OECD)
library(Quandl)
Quandl.api_key("YOU_KEY") 
library(eurostat)



```



```{r}
CLI_america <- c( "USALOLITOAASTSAM","USALORSGPNOSTSAM")%>% tq_get(get  = "economic.data",
                                                                   from = "1900-01-01",
                                                                   to   = "2023-12-31")   %>% rename ( Date = date, Value = price) 

CLI_america %>%
  mutate(symbol = ifelse(symbol == "USALOLITOAASTSAM", "CLI_USA", "GDP_USA")) -> CLI_america




CLI_america %>% spread( symbol , Value  ) %>%       filter(Date >= "2015-01-01")  %>% tail()

CLI_america %>% filter(Date >= "2015-01-01") %>% ggplot(aes(x=Date,y=Value,color=symbol))+
  geom_line(size=1)+
  geom_hline(yintercept = 100, color = palette_light()[[1]]) +
  facet_wrap(~ symbol, ncol = 2, scales = "free_y") +
  ggtitle("USA OECD Economic Cycle Road Map")+
  theme_tq() + 
  theme(axis.text.x = element_text(angle = 90, hjust = 1),
        axis.title.x = element_blank())+
  scale_color_tq()+
  scale_x_date(date_breaks = "3 months",date_labels = "%d (%b %y)") 

CLI_america %>% filter(Date >= "2000-01-01") %>% ggplot(aes(x=Date,y=Value, color = symbol))+
  geom_line(size=1)+
  geom_hline(yintercept = 100, color = palette_light()[[1]]) +
  ggtitle("USA OECD Economic Cycle Road Map")+
  theme_tq() + 
  theme(axis.text.x = element_text(angle = 90, hjust = 1),
        axis.title.x = element_blank())+
  scale_color_tq()+
  
  scale_x_date(date_breaks = "6 months",date_labels = "%y (%b)") 


```


```{r}
## Monthly Economic indicator for G7 Economies

dataset_list <- get_datasets()
search_dataset("MEI", data = dataset_list)
datasetG7 <- "MEI"
dstrucG7 <- get_data_structure(datasetG7)


G7_filter_list <- list("G-7", "LOLITOAA" )
G7_cli <- get_dataset(dataset = datasetG7 , filter = G7_filter_list )

G7_cli %>% tail(10) 
G7_cli$obsTime %>% AsDate() %>% ceiling_date( "month")  -1 -> G7_cli$obsTime
G7_cli %>% select(obsTime,obsValue) %>% rename(Date = obsTime, G7cli = obsValue) -> G7_cli
G7_cli %>%  filter(Date >= "2012-01-01") %>% ggplot(aes(x=Date,y=G7cli))+
  geom_line(size=1,color='red')+
  geom_hline(yintercept = 100, color = palette_light()[[1]]) +
  ggtitle("G7 OECD Economic Cycle Road Map")+
  theme_tq() + 
  theme(axis.text.x = element_text(angle = 90, hjust = 1),
        axis.title.x = element_blank())+
  scale_color_tq()+
  scale_x_date(date_breaks = "3 months",date_labels = "%d (%b %y)")












```
```{r}
## Inflation data


search_dataset("CPI", data = dataset_list)
dataset_cpi <- "PRICES_CPI"

filter_list <- list("G-7", "CPALTT01", "GY", "M" )
cpi_g7 <- get_dataset(dataset = dataset_cpi, filter = filter_list)
cpi_g7$obsTime %>% AsDate() %>% ceiling_date( "month")  -1 -> cpi_g7$obsTime
cpi_g7 %>% select(obsTime,obsValue) %>% rename(Date = obsTime, G7cpi = obsValue) -> cpi_g7
cpi_g7 %>% tail(10)
cpi_g7%>%  filter(Date >= "2000-01-01") %>% ggplot(aes(x=Date,y=G7cpi))+
  geom_line(size=1,color='red')+
  geom_hline(yintercept = 0, color = palette_light()[[1]]) +
  ggtitle("G7 CPI YoY %")+
  theme_tq() + 
  theme(axis.text.x = element_text(angle = 90, hjust = 1),
        axis.title.x = element_blank())+
  scale_color_tq()+
scale_x_date(date_breaks = "6 months",date_labels = "%d (%b %y)")

```

```{r}

## VIX

vix <- tq_get(c("^VIX"),
              get  = "stock.prices",
              from = "1900-01-01",
              to   = Sys.Date()) %>% 
  mutate(symbol = "VIX")

vix %>% 
  #filter(date > "2019-12-31") %>%
  ggplot(aes(x = date, y = adjusted, color = symbol)) +
  
  geom_line(color = "#E69F00") +
  geom_hline(yintercept = vix$adjusted %>% quantile(0.25,na.rm = T) , color = "blue" , linetype = "dotted") +
  geom_hline(yintercept = vix$adjusted %>% quantile(0.50,na.rm = T), color = "blue" , linetype = "dotted") +
  geom_hline(yintercept = vix$adjusted %>% mean() , color = "black" , linetype = "dashed") +
  geom_hline(yintercept = vix$adjusted %>% quantile(0.75,na.rm = T), color = "blue" , linetype = "dotted") +
  geom_hline(yintercept = vix$adjusted %>% quantile(1,na.rm = T), color = "blue" , linetype = "dotted")+
  #geom_smooth(color = "black") +
  scale_x_date(date_breaks =  "6 months", date_labels = "%d %b %Y", expand = c(0, 0)) +
  scale_y_continuous(breaks = seq(0,100, by = 5)) +
  scale_fill_brewer(type = "qual", palette =   "Set3", guide = F) +
  labs(title = "CBOE Volatility Index (VIX)", subtitle = "",
       x = "", y = "value", caption = "Data: Yahoo Finance") +
  theme_tq() + theme(axis.text.x = element_text(angle = 90, hjust = 1),
                     axis.title.x = element_blank()) +
  theme(legend.position = "none")+
  geom_ma(ma_fun = SMA, n = 30, color="red") +                
  geom_ma(ma_fun = SMA, n = 90, color="black") +
  
  coord_x_date(xlim = c("2015-07-01", "2023-12-31"))

vix %>% select(date,adjusted)%>%rename(Date = date, VIX = adjusted)-> VIX
VIX %>% tq_transmute(select = VIX , mutate_fun = to.monthly, indexAt = "lastof") -> VIX



```

```{r}

# Yields Spreads


TED_spread <- tq_get(c("TEDRATE"),
                     get  = "economic.data",
                     from = "1900-01-01",
                     to   = Sys.Date())
TED_spread %>% select(-symbol) %>% rename(Date = date, TED = price) -> TED_spread
TED_spread %>% tq_transmute(select = TED , mutate_fun = to.monthly, indexAt = "lastof") -> TED_spread


Tenyearminustwo_spread <- tq_get(c("T10Y2Y"),
                     get  = "economic.data",
                     from = "1900-01-01",
                     to   = Sys.Date())
Tenyearminustwo_spread %>% select(-symbol) %>% rename(Date = date, TENTWO = price) -> Tenyearminustwo_spread
Tenyearminustwo_spread %>% tq_transmute(select = TENTWO , mutate_fun = to.monthly, indexAt = "lastof") ->  Tenyearminustwo_spread

Tenyear <- tq_get(c("DGS10"),
                                 get  = "economic.data",
                                 from = "1900-01-01",
                                 to   = Sys.Date())
Tenyear %>% select(-symbol) %>% rename(Date = date, TEN = price) -> Tenyear
Tenyear %>% tq_transmute(select = TEN , mutate_fun = to.monthly, indexAt = "lastof") ->  Tenyear

Twoyear <- tq_get(c("DGS2"),
                  get  = "economic.data",
                  from = "1900-01-01",
                  to   = Sys.Date())
Twoyear %>% select(-symbol) %>% rename(Date = date, TWO = price) -> Twoyear
Twoyear %>% tq_transmute(select = TWO , mutate_fun = to.monthly, indexAt = "lastof") ->  Twoyear



```

```{r}

Tenyearminustwo_spread %>% 
 
  ggplot(aes(x = Date, y = TENTWO )) +
  
  geom_line(color = "#E69F00") +
  scale_x_date(date_breaks =  "2 years", date_labels = "%d %b %Y", expand = c(0, 0)) +
  scale_y_continuous(breaks = seq(0,100, by = 5)) +
  scale_fill_brewer(type = "qual", palette =   "Set3", guide = F) +
  labs(title = "10 Years minus 2 Years Yield spread", subtitle = "",
       x = "", y = "value", caption = "Data: Fred") +
  theme_tq() + theme(axis.text.x = element_text(angle = 90, hjust = 1),
                     axis.title.x = element_blank()) +
  theme(legend.position = "none")+
  geom_ma(ma_fun = SMA, n = 30, color="red") +                
  geom_ma(ma_fun = SMA, n = 90, color="black") +
  
  coord_x_date(xlim = c("1980-01-01", "2023-12-31"))


```



















```{r}
## PMI data

Quandl("ISM/MAN_PMI") -> PMI 

PMI  %>% as.tibble() %>%
  tq_transmute(select = PMI, mutate_fun = to.monthly, indexAt = "lastof") -> PMI

PMI %>% 
  ggplot(aes(x = Date, y = PMI)) +
  
  geom_line(color = "#E69F00") +
  geom_hline(yintercept = 50 , color = "black" , linetype = "dashed") +
  scale_x_date(date_breaks = "6 months", date_labels = "%d %b %Y", expand = c(0, 0)) +
  scale_y_continuous(breaks = seq( 30,70, by = 5)) +
  scale_fill_brewer(type = "qual", palette =   "Set3", guide = F) +
  labs(title = "PMI", subtitle = "",
       x = "", y = "value", caption = "") +
  theme_tq() + theme(axis.text.x = element_text(angle = 90, hjust = 1),
                     axis.title.x = element_blank()) +
  theme(legend.position = "none")+
  geom_ma(ma_fun = SMA, n = 3, color="red") +                
  geom_ma(ma_fun = SMA, n = 12, color="black") +
coord_x_date(xlim = c("2007-01-01", "2023-12-31"))
```



```{r}

## FED Funds data

FEDFUNDS <- tq_get(c("FEDFUNDS"),
                      get  = "economic.data",
                      from = "1900-01-01",
                      to   = Sys.Date())
    FEDFUNDS %>% select(-symbol) %>% rename(Date = date, FF = price) -> FEDFUNDS
    FEDFUNDS %>% tq_transmute(select = FF , mutate_fun = to.monthly, indexAt = "lastof") ->  FEDFUNDS



```

```{r}
## Join macro data

left_join( G7_cli, cpi_g7 , by = 'Date') %>%  left_join( VIX , by = 'Date')%>%
  left_join( Tenyearminustwo_spread, by = 'Date')%>%
   left_join( Tenyear , by = 'Date')%>% left_join( Twoyear  , by = 'Date')  %>% 
  left_join( FEDFUNDS  , by = 'Date') -> big_matrix



big_matrix %>% na.omit()   %>% mutate( G7cpi_m3 = SMA(G7cpi,3)       ) %>% 
  mutate( G7cpi_m36 = SMA(G7cpi,36) ) %>% mutate( G7cli_m3 = SMA(G7cli,3) ) -> big_matrix
big_matrix %>% tail(10)





```
```{r}

### Download etf prices and filter out etf quoted after 2015/04/30


etfs <- read_excel("C:/Users/Utente/Downloads/asd.xlsx")

prices_factors <-  etfs %>% 
  tq_get(get = "stock.prices", from = "1960-01-01") %>% 
  group_by(Ticker, Name)
prices_factors %>%  filter( min(date) <= "2015-04-30"   ) -> prices_factors

prices_factors

prices_factors %>% tq_transmute(select = close , mutate_fun = periodReturn , type = 'arithmetic', period = 'monthly' ,
                                indexAt ="lastof", values_fill = 0,
                                col_rename = "Returns") -> Returns_factors_etfs
Returns_factors_etfs %>% rename(Date = date) -> Returns_factors_etfs







```

```{r}

## Join macro and etfs returns data

left_join(  Returns_factors_etfs , big_matrix , by = 'Date' ) %>% group_by(Ticker, Name) -> Returns_MSCI_cli

Returns_MSCI_cli %>% tail(10)




```
```{r, fig.width = 30 ,fig.height = 30}
### performance when economy is in recovery after recession

Returns_MSCI_cli %>% filter ( G7cli < 100 & G7cli > lag(G7cli) ) %>% tq_performance( Ra = Returns , performance_fun =  table.Stats , Rb = NULL ) %>% arrange(desc( GeometricMean))


```

```{r,    fig.width = 30 ,fig.height = 30, dpi = 500, fig.retina = 2}



Returns_MSCI_cli %>%  filter( Date >= "2020-01-01"   ) %>% 
  filter( G7cli < 100 & G7cli > lag(G7cli) ) %>%
  ggplot(aes(y = Returns , x = reorder(Name, desc(Name)), color = Name)) +
  geom_hline(yintercept = 0, color = "red") +
  geom_boxplot(color = "black", alpha = 0.5) +
  geom_point() +
  scale_y_continuous(labels = scales::percent, limits = c(-0.5, 0.5)) +
  coord_flip() +
  labs(title = "Price returns of selected stocks/ETFs", subtitle = "2020-2023, monthly returns, ( G7cli < 100 & G7cli > lag(G7cli) )",
       x = "Stock", y = "return") +
  theme_tq() +
  theme(legend.position = "none")


```

```{r}


### performance when economy is in expansion and high inflation

Returns_MSCI_cli %>% filter ( G7cli > 100 & G7cli > lag(G7cli) & G7cpi > FF & G7cpi > G7cpi_m3 ) %>% tq_performance( Ra = Returns , performance_fun =  table.Stats , Rb = NULL ) %>% arrange(desc( GeometricMean))



```


```{r,  fig.width = 30 ,fig.height = 30, dpi = 500, fig.retina = 2}



Returns_MSCI_cli %>%  filter( Date >= "2020-01-01"   ) %>% 
  filter(G7cli > 100 & G7cli > lag(G7cli) & G7cpi > FF & G7cpi > G7cpi_m3 ) %>%
  ggplot(aes(y = Returns , x = reorder(Name, desc(Name)), color = Name)) +
  geom_hline(yintercept = 0, color = "red") +
  geom_boxplot(color = "black", alpha = 0.5) +
  geom_point() +
  scale_y_continuous(labels = scales::percent, limits = c(-0.5, 0.5)) +
  coord_flip() +
  labs(title = "Price returns of selected stocks/ETFs", subtitle = "2020-2023, monthly returns, ( G7cli < 100 & G7cli > lag(G7cli) )",
       x = "Stock", y = "return") +
  theme_tq() +
  theme(legend.position = "none")

```


```{r}

### performance when economy is in Slowdown ,inflation and rising interest rates

Returns_MSCI_cli %>% filter ( G7cli > 100 & G7cli < lag(G7cli) & FF > lag(FF) & G7cpi > G7cpi_m3 ) %>% tq_performance( Ra = Returns , performance_fun =  table.Stats , Rb = NULL ) %>% arrange(desc( GeometricMean))

```

```{r,  fig.width = 30 ,fig.height = 30, dpi = 500, fig.retina = 2}



Returns_MSCI_cli %>%  filter( Date >= "2020-01-01"   ) %>% 
  filter( G7cli > 100 & G7cli < lag(G7cli) & FF > lag(FF) & G7cpi > G7cpi_m3 ) %>%
  ggplot(aes(y = Returns , x = reorder(Name, desc(Name)), color = Name)) +
  geom_hline(yintercept = 0, color = "red") +
  geom_boxplot(color = "black", alpha = 0.5) +
  geom_point() +
  scale_y_continuous(labels = scales::percent, limits = c(-0.5, 0.5)) +
  coord_flip() +
  labs(title = "Price returns of selected stocks/ETFs", subtitle = "2020-2023, monthly returns, ( G7cli < 100 & G7cli > lag(G7cli) )",
       x = "Stock", y = "return") +
  theme_tq() +
  theme(legend.position = "none")




```