Challenging Japanese Stock Swing Trading with ChatGPT #44

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 in Japanese stock trading using AI.

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.

A loss of 19,480 yen on Shin-Etsu Chemical and TDK. The risk of chasing high prices that I was wary of has become a reality.

The realized loss on October 29 was -19,480 yen. Both Shin-Etsu Chemical Co., Ltd. and TDK, which I purchased the previous day, resulted in losses.

What I want to focus on this time is that even before purchasing, ChatGPT was wary of overheating in both stocks. Even if the evaluation of performance and materials is correct, if the purchase price and timing are wrong, it will not lead to profit the next day. That problem has been highlighted once again.

1. Sales results for October 29

Shin-Etsu Chemical Co., Ltd. (4063)

• Purchased on October 28: 6,066 yen x 100 shares

• Sold on October 29: 5,955 yen x 100 shares

• Realized loss: -11,100 yen

TDK (6762)

• Purchased on October 28: 2,768.8 yen x 100 shares

• Sold on October 29: 2,685 yen x 100 shares

• Realized loss: -8,380 yen

The total loss for the two stocks is 19,480 yen. All amounts are before deducting taxes, commissions, etc.

2. What did the previous day’s ChatGPT analysis misjudge?

Shin-Etsu Chemical Co., Ltd. announced its interim financial results on October 27, with operating profit increasing by 15.6% year-on-year. It rose with volume on the 28th, and ChatGPT also evaluated the financial results.

However, the momentum slowed after it rose to 6,115 yen in the morning session. ChatGPT set the recovery of the VWAP (Volume Weighted Average Price) as a purchase condition and warned against following the buy above 6,060 yen. The actual purchase price was 6,066 yen, which exceeds this warning line.

The same goes for TDK. It was bought on the back of materials such as an upward revision of the target stock price, and rose more than 4% in the morning session on the 28th. However, because it had risen significantly from the opening, it was excluded from the candidates in the initial screening due to overheating.

In the subsequent re-evaluation, it became a conditional purchase candidate based on the recovery to near the VWAP, but the selling price on the 29th was significantly lower than the purchase price.

For both stocks, the issue was not the evaluation of the news itself, but rather the judgment of the time horizon regarding whether the rise would continue until the next day. Especially with Shin-Etsu Chemical, the fact that it failed to adhere to the price ceiling indicated by ChatGPT itself is also a point for reflection.

3. Purchased 3 new stocks on October 29th

After confirming the losses, I purchased 100 shares each of the following 3 stocks.

Chugai Pharmaceutical (4519): 6,651 yen

While Core operating profit increased in the financial results, the stock price recovered after being sold off. It finished the morning session about 0.8% above the VWAP, making it a conditional buy candidate that evaluated both the earnings news and the improvement in price action. However, the purchase price is close to the morning high, and the risk of chasing the upside remains.

Capcom (9697): 3,731 yen

Despite good earnings, the stock plunged on the 28th due to disappointment over the full-year forecast remaining unchanged. The decline continued into the morning session on the 29th. The RSI was at a level indicating oversold conditions, but a reversal of the downward trend has not been confirmed. This purchase was a contrarian trade and was not one of ChatGPT’s primary recommended buy stocks.

Nomura Holdings (8604): 1,455.4 yen

The news is the increase in first-half profit and the dividend hike to 35 yen for the interim dividend. It recovered from its lows in the morning session on the 29th and finished above the VWAP. It was a conditional buy candidate that prioritized the continuity of the earnings news and the recovery of the stock price.

The total purchase amount for the 3 stocks is 1,183,740 yen. I will check the results during the next lunch break.

4. Cumulative 117 trades. What to verify next

With this, the total is 117 trades, with 62 wins and 55 losses. The win rate is 52.99%, but the cumulative realized profit/loss is -162,480 yen.

I need to verify why losses are accumulating even though the win rate is over 50%, looking at both purchase timing and the range of profit and loss.

Next time, I will compare the trend-following trades of Chugai Pharmaceutical and Nomura Holdings, which were based on earnings news, with the contrarian trade of Capcom, which plunged. Specifically, I will check whether a recovery above the VWAP leads to continued gains the next day, and at what stage one can expect a rebound for oversold stocks.

Thoughts on this time

・I have set up a scheduled task for selecting main stocks and a scheduled task for finding short-selling candidates, but I ended up buying Capcom, which was selected as the top candidate for short selling. It’s for a flimsy reason: I lost for the first time in a while, so I thought, ‘Trend reversal starting today! Let’s go with a contrarian trade!’ ・It’s pretty discouraging to have this week’s winnings (though it’s only 3 days’ worth) wiped out. Even though it’s obvious that I can’t win all the time.

※This article is a record of personal trades and verification, and does not recommend buying or selling specific stocks.

#Investment #StockInvestment #JapaneseStocks #Stocks #AssetManagement #AssetFormation #SwingTrade #InvestmentRecord #IndividualInvestor #StockInvestmentBeginner #ChatGPT #AIInvestment #DayTrade #InvestmentBeginner #StockMarket