/updiv Command

/divrank Command

/news Command

/trend Command

The trend command computes the relevant metrics for trend analysis of a security in the given lookback period. This aspect caters to the technical minds who are keen in studying the trendlines and price movement of stocks.

To use this command, you can key in "/trend" along with the stock code of the security and the number of years as lookback period.

To demonstrate using DBS stock, you can key in "/trend D05, 10" to analyze the trend data for DBS over the past 10 years.

The mean refers to the average daily stock price for the defined period (i.e. 10 years in this case). Visually, you can imagine a horizontal line cutting across the chart at 16.06. 

On top of that, users are able to create their own linear regression charts for trend analysis. 

The stock trend line refers to a linear regression line (or line of best fit) drawn through the daily stock price data. Users are able to use the long-term trend line as a gauge to determine the valuation of the security at the current price (ie. if a stock is trading above or below the regression trendline).

The inclusion of 4 standard deviation lines serves as additional metrics to quantify the degree of deviation from the main regression trendline. For example, the stock price of DBS has reached 1 standard deviation above the trendline in 2018 and fallen below 1 standard deviation from the trendline in 2016 and 2020. 

The price data for these 5 metrics represent the latest price points at the end of the chart. 

Naturally, this command offers an over-simplistic approach in determining the valuation of a particular security. Exercising the appropriate due diligence is definitely recommended. 

/support Command

Last but not least, if you have found SGStocksBot to be useful and would like to show your appreciation, the support command allows the user to contribute and help support the running of the bot via PayLah payment link.

One last thing to note is that the bot will idle and go to sleep after 30 minutes of inactivity. So do expect the bot to take a few seconds to respond to your command if it has been more than 30 minutes since your last interaction with it. Other than that, the bot should respond almost instantaneously.

If time permits, I may roll out additional bot features/commands in the future. In the meantime, I hope you enjoy the bot and please let me know of any feedback that I can improve on it. 

If you do not wish to miss out on any articles, you may consider following the facebook page for timely update. 

Disclaimer: Kindly note that the content is published for general information purpose only. The information presented should not be used or construed as an offer to sell, a solicitation of an offer to buy, a recommendation for any security mentioned or as professional financial investment advice.

from pandas_datareader import data as pdr
import pandas as pd
import yfinance as yf

yf.pdr_override()

# Indicate start and end date
start = '1987-01-01'
end   = '2016-01-01'

# Generate STI price data
sti = pdr.get_data_yahoo('^STI', start, end)['Adj Close']

# Extract STI year-end price data
yearly_sti = sti.groupby(pd.Grouper(freq='Y')).tail(1)

# Generate STI annual returns
yearly_rtn = round(yearly_sti.pct_change()*100,2)

# List the election years
election_year = [1988, 1991, 1997, 2001, 2006, 2011, 2015]

# Filter STI returns on election years
election_year_rtn = []
for year in election_year:
    election_year_rtn.append(float(yearly_rtn.values[yearly_rtn.index.year == year]))

# List the election dates
election_dates = pd.to_datetime(['1988-09-03','1991-08-31','1997-01-02',
                                 '2001-11-3','2006-5-6','2011-5-7','2015-9-11'])

# Obtain the dates one month before and after the Polling Day
election_1m_before = []
election_1m_after = []

for date in election_dates:
    election_1m_before.append(date - pd.DateOffset(months=1))
    election_1m_after.append(date + pd.DateOffset(months=1))

# Adjustment if the dates coincide with the weekends    
election_1m_before_adj= []
election_1m_after_adj = []
counter1 = 0
counter2 = 0

for item1 in election_1m_before:
    if not (sti.index == item1).any():
        election_1m_before_adj.append(election_dates[counter1] 
                                      - pd.DateOffset(months=1,days=1))
    else:
        election_1m_before_adj.append(item1)
    counter1+=1
    
for item2 in election_1m_after:
    if not (sti.index == item2).any():
        election_1m_after_adj.append(election_dates[counter2] 
                                     + pd.DateOffset(months=1,days=1))
    else:
        election_1m_after_adj.append(item2)
    counter2+=1

# Compute STI returns one month before and after the Polling Day
diff = sti.loc[election_1m_after_adj].values - sti.loc[election_1m_before_adj].values
returns = diff/sti.loc[election_1m_before_adj].values * 100

from scipy.stats import pearsonr 

returns = list(returns)

# List the vote percentage won by PAP for the past elections
vote = [63.2, 61, 65, 75.3, 66.6, 60.14, 69.86]

# Calculate the correlation coefficient between the vote count and STI returns
corr, _ = pearsonr(vote, returns) 
print('Pearson correlation: %.3f' % corr)

# Drop the element for 2015 General Election
returns.pop()
vote.pop()

# Calculate the correlation coefficient between the vote count and STI returns
corr, _ = pearsonr(vote, returns) 
print('Pearson correlation: %.3f' % corr)

from pandas_datareader import data as pdr
import pandas as pd
import yfinance as yf

yf.pdr_override()

# Indicate start and end date
start = '2000-01-01'
end   = '2020-01-01'

# Generate STI price data
sti = pdr.get_data_yahoo('^STI', start, end)['Adj Close']

# Calculate STI daily returns
sti_returns = sti.pct_change() 

import quantstats as qs

# Generate Monthly Heatmap for STI
qs.plots.monthly_heatmap(sti_returns)

# Calculate the average returns for each month
sti_monthly_returns = qs.stats.monthly_returns(sti_returns).iloc[:,:-1]
sti_monthly_returns.mean()*100

import statistics 

# Run a loop and calculate the standard deviation of STI returns by month
sd = {}
for month in sti_monthly_returns:
    sd[month] = statistics.stdev([returns for returns in sti_monthly_returns[month]])

pd.DataFrame.from_dict(sd, orient='index',columns=["Standard Deviation"])

# Set 2 empty lists
positive_month = []
negative_month = []

# Run a loop and count the number of years with positive/negative STI returns by month
for month in sti_monthly_returns:
    positive_count = 0 
    negative_count = 0
    for returns in sti_monthly_returns[month]:
        if returns >= 0:
            positive_count += 1
        else:
            negative_count += 1
    positive_month.append(positive_count)
    negative_month.append(negative_count)