Challenging Japanese Stock Swing Trading with ChatGPT #34
I continue to engage in short-term swing trading, where I purchase Japanese stocks during my lunch break on weekdays and, in principle, sell them during my lunch break on the next business day. I use ChatGPT for stock selection and trading decisions, verifying with my own money whether it is actually possible to make a profit using AI for Japanese stock trading.
This article, excluding the final “Thoughts on this time” section, was also created by ChatGPT. The main text is organized based on actual trading records and the analysis performed at that time.
All 3 stocks fell, a loss of 58,800 yen in one day. Also attempting margin short selling
October 15th was a difficult market, with the Nikkei Stock Average falling by more than 3,000 yen from the previous day. All three stocks I purchased the day before resulted in losses, totaling a negative 58,800 yen.
In this trade, in addition to the conventional swing trading that starts with buying, I have newly incorporated “margin short selling,” which aims for profit from stock price declines.
Selling results for October 15th
The results for the three stocks purchased on October 14th and sold on the 15th are as follows.
Sony Group (6758)
purchased 100 shares at 3,861.9 yen and sold at 3,855 yen. The profit/loss was **-690 yen**.
Nippon Yusen (9101)
purchased 100 shares at 7,554 yen and sold at 7,294.8 yen. The profit/loss was **-25,920 yen**.
Mitsubishi Electric (6503)
purchased 100 shares at 5,240.9 yen and sold at 4,919 yen. The profit/loss was **-32,190 yen**.
The total realized profit/loss for the three stocks is -58,800 yen (excluding taxes, commissions, etc.). The declines in Mitsubishi Electric and Nippon Yusen, in particular, increased the loss.
Was ChatGPT’s judgment from the previous day correct?
In the analysis on October 14th, I evaluated Nippon Yusen for being above the VWAP and recovering from the morning decline. Mitsubishi Electric was also judged as a “conditional buy” based on the raising of highs and lows and its movement above the VWAP.
However, both fell significantly the next day. The expectation that the strength of the morning session would continue until the next day was wrong, and it is a result of not being able to respond to sudden changes in the entire market with only short-term upward shapes.
On the other hand, Sony Group lacked momentum as of the previous day, and ChatGPT’s judgment was “WATCH.” Although I actually purchased it, the decline the next day was only about 0.18%, making it the smallest loss among the three stocks.
The results this time highlight the challenge of how to incorporate the risk of the entire market falling sharply the next day, not just the technical evaluation of individual stocks.
New purchases are Kao and KDDI. Moving to defensive stocks
On October 15th, while the entire market was falling significantly, I focused on daily necessities and telecommunications, which tend to have relatively stable price movements.
The new purchases were 100 shares of Kao (4452) at 3,702 yen and 100 shares of KDDI (9433) at 2,902.5 yen.
In ChatGPT’s preliminary analysis, Kao was evaluated for the shift of funds to daily necessities and its relative strength against the overall market, and KDDI was evaluated for the firmness of telecommunications stocks and its price movement near the VWAP.
I will verify whether this purchase was ultimately correct based on the price movement on the next business day.
Introducing margin short selling. Can I aim for profits even in a falling market?
Another new attempt is short selling using margin trading.
On October 15th, I newly short-sold 100 shares of Socionext (6526) at 1,840.3 yen.
As of the morning session, the stock price was 1,847 yen, down 6.67% from the previous day, and about 1.56% below the VWAP. While ChatGPT evaluated the downward momentum, it also pointed out the risk of a rebound after a sharp drop.
While margin short selling allows for profit-seeking even in a falling market, it carries risks of losses from sharp rebounds and costs such as stock lending fees. In this trade, I will verify whether the downward trend continues until the next day and whether the short-selling price was appropriate.
As of October 15th, the cumulative total is 93 trades, with 45 wins, 46 losses, and 2 pending results. The cumulative realized profit/loss is **-190,260 yen**, and for October alone, it is **-161,900 yen** (excluding taxes, fees, etc.).
Moving forward, I will judge new buys during market plunges more strictly and check if adding margin short selling leads to an improvement in profit and loss.
My thoughts on this time
・Oh no! Oh no! Oh no!
・That said, there is nothing to be done about today’s market, so it is what it is. Chappy might be a bit of a klutz, but maybe it’s good at picking stocks that will fall, so I decided to try margin short selling. I was thinking I needed to open a margin account, but it seems I had already opened one. For now, I chose Socionext as a semiconductor stock with a small amount. It said that the decline in semiconductor stocks was small and that they would fall further.
・For swing trades where you sell at noon the next day, it seems the fee is only about 0.015% of the selling price, so margin trading is surprisingly viable in a falling market. I am not using leverage.
・Safely? It looks like semiconductor stocks are falling in the US market.
※This article is a record of personal trades and verification, and does not recommend buying or selling any specific stock.
#Investment #StockInvestment #JapaneseStocks #Stocks #AssetManagement #AssetFormation #SwingTrading #InvestmentRecord #IndividualInvestor #StockInvestmentBeginner #ChatGPT #AIInvestment #MarginTrading #DayTrading #StockBeginner