Deepmoney 34b 200k Base

🚀 Deepmoney

Deepmoney is a financial model aiming to offer more effective investment decision - making support in the ever - changing financial market.

Sft Versions:

67b Chat 34b Chat

Introducing Greed in the Seven Deadly Sins series of models.

• Full - para pre - training on Yi - 34b
• High - quality research reports
• High - end cleaning process

🚀 Quick Start

1. What do I want to do?

Most of the current so - called financial models are mainly trained on public knowledge. However, in the actual financial field, this public knowledge is often seriously insufficient for interpreting the current market. If you're interested, you can learn about the various propositions of Keynes, Friedman, and even current behavioral finance. According to my observation, most financial models can't make investment judgments because they are trained on ordinary textbooks, entry - level analyst exams, and even company public reports. I think this has very little value for investment.

You may think I'm joking, but the fact is that the logic of many subjective analysts may not be as rigorous as that of large models of 34b and above (excluding the excellent ones, of course). The market is changing every moment, with a large amount of real - time news and massive data. For most retail investors, instead of waiting for a lousy analyst to write a report, why not use a large model to make a pipeline?

In my plan, this model is the base model of this process. In my plan, models such as information collector, target judge, qualitative analyst, quantitative analyst, and data extractor are all part of this process. But it's undoubtedly important for the model itself to master a large number of qualitative and quantitative methods. That's why this model was born.

2. About the data

As I just said, a lot of public knowledge has some questionable validity - but that doesn't mean it's wrong. The theoretical support behind many research methods in research reports also relies on this knowledge. So in my training, I picked up some college textbooks and some professional books. The quantity is not large, but the quality is good. In addition, I selected a large number of research report data from 2019 to December 2023. These reports are issued by a variety of publishers, including traditional brokers and professional research institutions. Most of them are paid and only available to institutions. But I got them anyway through various means.

If you have read research reports, especially high - quality ones, you'll find that research reports are all about subjective judgment + quantitative analysis, and data support in quantitative analysis is crucial to the entire logical chain. In order to extract this data (most of them are in the form of graphs or tables), I tried a lot of multi - modal models, and the process was very painful. The conclusion is that cog - agent and emu2 are very effective for this kind of tasks. In order to better extract information, I created a process that summarizes the context of research reports as part of the prompt.

Finally, I made a blend of the data. General data is not included because it is just for greed. Moreover, the knowledge in industry research reports is comprehensive enough.

3. About training

Raw text, full parameter training. The base uses long context yi - 34b - 200k. This is necessary to complete and understand an in - depth report.

Of course, I also did a sft. This is the analyzer in my process – I haven’t broken down the qualitative and quantitative analysis yet, but I’m already blown away by how well it works.

4. Evals

Let's sample some recent events, simulate the event - driven securities analysis process in the real world, and conduct a comparative test against deepmoney and gpt4. Because the impact of events on the market is relatively invisible, it's difficult to evaluate the effect without a rigorous back - testing process. And our output needs to be analyzed using many quantitative methods. So I post the results here so that everyone can make a perceptual evaluation of the production results. I have a global news crawling system, and there are a lot of news every moment.

In my process, I deduplicate and make subjective and objective judgments on these news. These can be solved by traditional BERT. Then for deepmoney, three steps are used to process the incoming news:

1. Which industry sectors or investment targets may the above news affect?
2. Please design a quantitative method to study the impact of the above news on the industry. And accordingly explain what data needs to be used.
3. Based on the following data, please design a set of specific quantitative methods to quantitatively analyze the impact of the above news on the industry.

Here is the three - step answer for deepmoney and gpt4:

Deepmoney:

(1). 
Recently, British pharmaceutical company GlaxoSmithKline (GSK.US) announced that it will acquire Aiolos Bio, a US clinical - stage biopharmaceutical service provider, with a down payment of $1 billion and milestone payments of $400 million. GSK is targeting Aiolos Bio's only core pipeline - the monoclonal antibody AIO - 001 for treating adult asthma. The AIO - 001 pipeline is not self - developed by Aiolos Bio; it is licensed from China's leading pharmaceutical company, Hengrui Medicine. A pharmaceutical analyst said that overall, in this AIO - 001 monoclonal antibody transaction, the so - called "middleman profiting" statement may be media hype. Regarding the view of "Hengrui Medicine easily selling," he said: "One (Aiolos Bio) is taking a gamble, placing bets, and getting a big order once every few years; one (Hengrui Medicine) is a manufacturing business with overseas incremental business on the side. They are doing completely different businesses, and Hengrui is still the profitable one." (Pengpai News)
Which industry sectors or investment targets may the above news affect?

ANS:
The above news primarily affects the pharmaceutical, biotechnology, and healthcare industries as it involves an acquisition by GSK of Aiolos Bio, which is focused on developing treatments for asthma using a single monoclonal antibody called AIO - 001. The deal also has implications for companies in these sectors that could be involved in similar deals or collaborations to expand their pipelines and market presence.
******************************************************************************
(2). Recently, British pharmaceutical company GlaxoSmithKline (GSK.US) announced that it will acquire Aiolos Bio, a US clinical - stage biopharmaceutical service provider, with a down payment of $1 billion and milestone payments of $400 million. GSK is targeting Aiolos Bio's only core pipeline - the monoclonal antibody AIO - 001 for treating adult asthma. The AIO - 001 pipeline is not self - developed by Aiolos Bio; it is licensed from China's leading pharmaceutical company, Hengrui Medicine. A pharmaceutical analyst said that overall, in this AIO - 001 monoclonal antibody transaction, the so - called "middleman profiting" statement may be media hype. Regarding the view of "Hengrui Medicine easily selling," he said: "One (Aiolos Bio) is taking a gamble, placing bets, and getting a big order once every few years; one (Hengrui Medicine) is a manufacturing business with overseas incremental business on the side. They are doing completely different businesses, and Hengrui is still the profitable one." (Pengpai News)

Please design a quantitative method to study the impact of the above news on the pharmaceutical industry. And accordingly explain what data needs to be used.

ANS:
To quantitatively analyze the impact of this acquisition news on the pharmaceutical industry, we can use regression analysis or other statistical modeling techniques. The key is to identify independent variables that might influence stock prices in the pharmaceutical sector and then determine how much each variable affects changes in stock price.

Here's an overview of the steps involved in designing such a quantitative approach:

Step 1: Data Collection
Collect historical data for the period you wish to study (e.g., quarterly or yearly) from various sources like Bloomberg, Yahoo Finance, etc. Here's a list of some potential datasets needed:

* Stock market indices related to pharmaceutical industries both locally (e.g., S&P Pharmaceutical Index) and globally (e.g., S&P Global Healthcare Index)
* Financial performance indicators of major players within the same space as GSK/Aiolos Bio and those affected by their acquisitions
* News sentiment scores around similar deals in the past
* Macroeconomic factors like GDP growth rate, inflation rates, interest rates, etc.
* Industry specific factors like R&D spending trends, regulatory environment changes, patent expirations, etc.

Step 2: Model Selection
After collecting all relevant information, choose suitable econometric models depending upon the nature of relationships between dependent & independent variables. Ordinary Least Squares Regression would work if linearity assumption holds true; otherwise, Logit/Probit model could be considered too.

Step 3: Variable selection
Selecting appropriate explanatory variables after checking multicollinearity among them using VIF values. This will help us understand which variables have significant effects on stock returns.

Step 4: Hypothesis testing
Perform hypothesis tests like F - test, t - test, Durbin - Watson test, etc., to check whether our chosen regressors are statistically significant at different levels of significance.

Step 5: Coefficient interpretation
Interpret coefficients obtained from Step 4 while keeping in mind signs and magnitudes must align with economic theory. Positive coefficient means positive relationship whereas negative implies inverse relation.

Step 6: Model evaluation
Evaluate goodness - of - fit measures like R^2, Adjusted R^2, Akaike Information Criterion(AIC), Bayesian Information Criterion(BIC), Root Mean Square Error(RMSE), Mean Absolute Deviation(MAD), etc., to ascertain model fitness.

Step 7: Forecasting
Use estimated parameters along with future exogenous variable values to predict future outcomes for any given time period.

In conclusion, this quantitative methodology aims to measure the effect of GSK's acquisition on the overall pharmaceutical industry by studying its impacts on stock prices and financial performances across several companies in the field.
****************************************************************************
(3). Recently, British pharmaceutical company GlaxoSmithKline (GSK.US) announced that it will acquire Aiolos Bio, a US clinical - stage biopharmaceutical service provider, with a down payment of $1 billion and milestone payments of $400 million. GSK is targeting Aiolos Bio's only core pipeline - the monoclonal antibody AIO - 001 for treating adult asthma. The AIO - 001 pipeline is not self - developed by Aiolos Bio; it is licensed from China's leading pharmaceutical company, Hengrui Medicine. A pharmaceutical analyst said that overall, in this AIO - 001 monoclonal antibody transaction, the so - called "middleman profiting" statement may be media hype. Regarding the view of "Hengrui Medicine easily selling," he said: "One (Aiolos Bio) is taking a gamble, placing bets, and getting a big order once every few years; one (Hengrui Medicine) is a manufacturing business with overseas incremental business on the side. They are doing completely different businesses, and Hengrui is still the profitable one." (Pengpai News)

Based on the following data, please design a specific set of quantitative methods to quantitatively analyze the impact of the above news on the pharmaceutical industry.

* Stock market indices related to pharmaceutical industries both locally (e.g., S&P Pharmaceutical Index) and globally (e.g., S&P Global Healthcare Index)
* Financial performance indicators of major players within the same space as GSK/Aiolos Bio and those affected by their acquisitions
* News sentiment scores around similar deals in the past
* Macroeconomic factors like GDP growth rate, inflation rates, interest rates, etc.
* Industry specific factors like R&D spending trends, regulatory environment changes, patent expirations, etc.

ANS:
To quantitatively analyze the impact of this news on the pharmaceutical industry, we can use several quantitative methods that involve analyzing stock prices, financial performance indicators, news sentiment scores, macroeconomic factors, and industry - specific factors. Here is an outline of how you might go about it:

1. Stock Market Indices Analysis: Monitor daily or weekly fluctuations in local (e.g., S&P Pharmaceutical Index) and global (e.g., S&P Global Healthcare Index) stock market indices related to the pharmaceutical industry before, during, and after the announcement of the acquisition. Analyze these fluctuations using statistical techniques such as regression analysis, ANOVA testing, or time series analysis to determine whether there were significant price movements due to this event.
2. Financial Performance Indicators: Examine the financial performance metrics of other companies operating in the same therapeutic area(s) as Aiolos Bio and GSK prior to the acquisition announcement. These could include revenue growth rates, net income margins, return on equity (ROE) ratios, earnings per share (EPS), etc. Then compare them with post - acquisition figures for any noticeable differences that may have resulted from the deal. You can also look at the overall profitability trend among firms within the same space to gauge its potential effect on investor confidence.
3. News Sentiment Scores: Collect media coverage surrounding previous mergers & acquisitions (M&As) involving biotech startups similar to Aiolos Bio over a certain period (e.g., 5 years). Use natural language processing tools or machine learning algorithms to extract positive/negative sentiments associated with each article. Compare these scores...

📄 License

This project is under the Apache - 2.0 license.