The Hong Kong stock market demonstrates strong resilience during the earnings season, with short-term market focus to be aware of!

Hong Kong stocks regained some ground after falling today, versus yesterday's U.S. stocks $Dow Jones Industrial Average(.DJI.US)$ $S&P 500 Index(.SPX.US)$ $Nasdaq Composite Index(.IXIC.US)$ The weak performance is surprising. One of the reasons that support this month's constant pointing to the good is that results are believed to be an important element. Except for today. $POP MART(09992.HK)$ und $LAOPU GOLD(06181.HK)$ The performance of the Consumer Stock Industry has been good, and the share value has been poor. eg $XPENG-W(09868.HK)$ und $XIAOMI-W(01810.HK)$ The performance of the market is expected, although the share price does not show an increase, but the development direction and profitability of the company indicate that conditions continue to be good in the long term. There are two main reasons for the rebound in Technology stocks, namely reducing the profitability improvement required by growth or regulatory model performance, or showing an unforeseen impact within the Industry (Market Overshoot).

There was no increase in share prices after the performance of Xiaomi Mi and Xiaobo, but the two showed the plight of the Cars Industry and no losses in the market ahead of time. Under extremely favorable conditions, Xiaomi Cars increased its sales volume further, and its gross margin increased to 26.2%. The G6 and G7 faced a slight downward pressure on prices this year, but the gross margin increased to 16.5% (G7 was released in July, without reflecting operating performance). In the early morning, the market had an overview of the Cars industry, $GEELY AUTO(00175.HK)$ The share price also rose to a 52-week high following the results, $LEAPMOTOR(09863.HK)$ Set a record high after the results.

Industry examples where the internal industry is under greater competitive pressure, such as platform consumption, and the launch of the industry is better in the short term. $TONGCHENGTRAVEL(00780.HK)$ After the results, the share price is stable; however, $JD-SW(09618.HK)$ After the results, the share price was pressured and hit new lows this year. $NTES-S(09999.HK)$ The expected performance, the share price reached a new low in the third quarter. From the perspective of the weak stock market, the gap for the temporary decline is not evident. It is believed that the main reasons and values are relevant. However, the average valuation of HKEx Technology shares is currently low and no less profitable reasons have been reflected in the share price.

Of course, when it comes to individual stock operations, investors are advised to keep an eye out for some performance dividends or strong stocks. The message, however, is that Hong Kong stocks are still showing good bearish resistance, and the pattern of at least waves above waves is maintained.

Whatever $Hang Seng Index(800000.HK)$ either $Hang Seng TECH Index(800700.HK)$ , The technical trend has remained high above the wave to date. At the same time, the Hang Seng Index, in particular, fell to the bottom of the channel today, reflecting the strength of the entire uptrend.

It is worth noting that the A shares have performed very strongly, $SSE Composite Index(000001.SH)$ und $Shenzhen Component Index(399001.SZ)$ Bullish gains, domestic insurance stocks $CPIC(02601.HK)$ $CHINA LIFE(02628.HK)$ $NCI(01336.HK)$ $PING AN(02318.HK)$ The trend has remained strong to this day. Capital Markets are in a good direction, and there can be no significant upside to shorting. Under the support of a series of fiscal policy, MMF policy or trade talks, the outlook for corporate profit growth is now on a higher footing.

Focus Beyond Performance

Apart from the difficult stock performance and the early announcements, Kuaishou will immediately announce the results of the public market, and one of the shareholders will be responsible for the overall performance of the shares that are not on the market.

There is not much more information to share and share with everyone, so the following are two reasons why it is worth paying attention to your parents:

HIBOR Recovery

The first focus was on HIBOR's significant drawdown, leading to $USD/HKD(USDHKD.FX)$ An increase in the ratio of 7.8, the change in the interest rate in the short term is mainly related to the change in the US interest rate forecast and the level of the set trade is flat, the market is expecting HIBOR to rise to 3cm to the left, and the rate of Hong Kong dollar continues to trend strongly against the US dollar, and only the difference is reflected in the short term. For the whole stock market or the Hang Seng Index, the impact will not be large, but there will be an exact reflection of the local property sector in the short term. As for the real estate market, there is no time for people to discuss, but one thing that can be certain is that property prices are not simply a factor in earnings, but sentiment and economic outlook are the most important factors in predicting the future.

More than a few local real estate stocks have seen a rebound today, with XIN real estate being the fifth place in the rise in the ConstituentStocks in Hang Seng Index. If short line HIBORs do not show further abstraction, funds may continue to flow to this Sector.

$SHK PPT(00016.HK)$ $HENDERSON LAND(00012.HK)$ $CK ASSET(01113.HK)$ $SINO LAND(00083.HK)$ $SWIREPROPERTIES(01972.HK)$ $LINK REIT(00823.HK)$

Duma Research Report on Semiconductors in China

The second most interesting is the upcoming mass study report. This Visitor's Share was featured at the end of the live broadcast of the Investment Talk (How to chart the key sectors ahead of Hong Kong stock declines? ), approximately after 53:00 during the live broadcast.

Simply put, this report studies the costs and benefits of data centers using different chips under the same reasonable assumption

Comparable scenario with 75 server enclosures in a medium-sized AI data center (note that the cabinet is not a mere chip)

Including chips that everyone is familiar with mainstream ( $NVIDIA(NVDA.US)$ 、 $Advanced Micro Devices(AMD.US)$ 、ASIC $Alphabet-C(GOOG.US)$ either $Amazon(AMZN.US)$Chips, and Huawei)

One of the focuses is TCO (total cost of ownership, the cost of building the entire data center), NVDA is the highest, since GB200 is the most expensive, followed by AMD, the cost of ASICs is lower. The performance of TCO is just a reference to the current chip price level.

Following the Run AI recommendation model after construction, the reference API fee was charged at $0.2 per 100 million Tokens, and 70% of utilization was calculated from left to right (Industry average at present = Average), resulting in lower profits

FIRST, AS EVERYONE KNOWS, NVDA'S CHIPS ARE GREAT, PROFITABILITY IS THE MOST EFFICIENT; SECOND IS GOOG'S 6TH GENERATION TPU, AND THIRD IS AWS' TR2. The success of Blackwell and ASIC is no stranger to the success of all.

The results below are more interesting. First of all, AMD's disappointing performance and data center operations will be the only loss, but this is not simply a reflection of the chip, it is about a scale-up. This means that in the current development of AI data centers, in addition to the importance of chips, cabinet platform schemes or scale-out technologies are also very important.

Speaking of which, this is the bright spot of the whole report. Hua's chipped chip strategy was successful with a 47.9% efficiency on a server platform connected to 384Q910C chip chains. This produces two points of reference:

— Domestic AI models and data centers are conditionally profitable

The chip-chip strategy is successful, but the price is that large chip displacement is required, without the need for large production

While the report may not have been the shock of deepseek's appearance in the early days, it undoubtedly led investors to believe that China is an important element of competitiveness on the AI investment track. In addition, other than the large AI model companies that are familiar with (e.g. $BABA-W(09988.HK)$ $TENCENT(00700.HK)$ $BIDU-SW(09888.HK)$ $SENSETIME-W(00020.HK)$ ), etc., market demand forecasts for AI infrastructure (e.g. semiconductors, fiber optics or cabinet technologies, etc.) are also increasing, e.g. $SMIC(00981.HK)$ $HUA HONG SEMI(01347.HK)$ $YOFC(06869.HK)$ $ZTE(00763.HK)$etc.