Predicting whether corporations are successful is crucial for conducting investment decisions and the design of effective economic policies. However, earlier research on high-growth corporations considered key to economic development-firmly showed low predictive accuracy, which suggests that growth may be largely random. Does this assumption still occur in the AI era, where huge amounts of data and advanced analytical methods are available? Can AI techniques overcome difficulties in anticipating high growth corporations? These questions were asked in chapter I co -authorized the co -author in who reviewed the scientific contribution in the company’s growth forecast using AI methods.
According to the definition of Eurostat-OOCD (Organization of Economic Cooperation and Development), High -height corporations Do corporations have at least 10 employees in the initial growth period and “average annual growth greater than 20% per year, within three years”. Growth may be measured by the number of company employees or its trade. The subset of high -growth corporations, referred to as “Gazelles”There are young companies-the start-ups-who are up to five years old and have rapid development.
Companies with high growth drive development, innovation and job creation. Identification of corporations with a high increase in potential enables investors, start-up incubators, accelerators, large corporations and decision-makers to adapt the potential investment opportunities, strategic partnerships and allocation of resources at an early stage. Forecasting results for start-ups is tougher than for large corporations due to limited historical data, high uncertainty and relying on qualitative aspects, comparable to the founder’s experience and market matching.
How random is strong growth?
Accurate growth forecasting is particularly necessary, taking into account the high start-up failure indicator. One in five start-ups fails in the first yr and Two -thirds fail in 10 years. Some start-ups may significantly contribute to creating jobs: research Data evaluation from Spanish and Russian corporations In the years 2010–2018 it showed that while Gazelles constituted only about 1-2% of all corporations in each countries, they were responsible for about 14% of employment in Russia and 9% in Spain.
Companies with high growth are “Commonly considered necessary to stimulate economic growth and employment” But they are difficult to discover. Interested parties need accurate growth forecasts to help optimize decision making and minimize risk by identifying corporations with the highest success potential.
To understand why some corporations grow faster than others, researchers looked at various aspects including entrepreneurs’ personality, competitive strategy, available resources, market conditions and macroeconomic environment. However, these aspects explained only a small part of the volatility of the company’s growth and were limited in their practical application. This led to the suggestion that predicting the development of recent corporations is like Playing a likelihood. Another point of view argued that the problem of growth forecasting may result from the methods used, suggesting “The illusion of randomness”.
Because the company’s growth is a complex, diverse, dynamic and non -linear process, taking a recent set of methods and approaches, comparable to those driven by Big Data and AI, can shed recent light on the debate of growth and forecasting.
AI offers recent opportunities to anticipate highly developing corporations
AI methods are increasingly accepted to forecast the company’s growth. For example, 70% of Venture Capital accepts AI To increase internal productivity and facilitate and speed up acquisition, screening, classification and monitoring of high potential start-ups. Crunchbase, a company’s data platform, claims that internal tests have shown that its AI models can predict the success of launching with launch “95% precision” Analyzing hundreds of signals. These development promise to fundamentally change the way investors and corporations approach decisions in private markets.
The benefits of AI techniques consist of their ability to process much higher volume, diversity and speed of data on corporations and their environments compared to traditional statistical methods. For example, machine learning methods comparable to Random forest (RF) and the least ruthless contraction and selection operator (Lasso) Help discover key variables affecting business results in data sets with a large number of predictors. AND Model of a large “molten” language It has been shown that it provides for the success of the start-up using each structured (organized in tables) basic information and unstructured (unorganized and more complex) text descriptions. AI techniques help increase the accuracy of the company’s growth forecasts, discover the most significant growth aspects and minimize human prejudices. As some scholars noted, a higher forecast indicates that perhaps The company’s growth is less random than previously thought. In addition, the ability to capture data in real time is particularly invaluable in fast, dynamic environments, comparable to advanced technology industries.
Challenges remain
Despite the rapid progress of AI, there is still significant development potential. Although forecasting high growth corporations has been improved thanks to modern AI techniques, studies Indicate that this is still a challenge. For example, the success of starting often depends on rapidly changing and intangible aspects that are not easily captured by data. Further methodological progress is beneficial, comparable to the inclusion of a wider scope of predictors, various data sources and more sophisticated algorithms.
One of the fundamental challenges for AI methods is their ability to provide explanations regarding the anticipated forecasts. Forecasts generated by complex models of deep learning resemble a “black box”, and the causal mechanisms that transform the input data into the output remain unclear. Producing more Explaining artificial intelligence He became one of the key goals set by the research community. Understanding what may be explained and what can’t be (yet) can’t be explained using AI methods, it may well higher conduct practitioners in identifying and supporting corporations with high increases.
While start-ups offer the potential of significant return on investment, they bear a significant risk, thanks to which the most significant are the exact selection and accurate prediction. As the AI models evolve, various and unstructured data sources and real -time signals can be more and more integrated to detect early potential success indicators. Progress is expected to increase scalability, accuracy, speed and transparency of AI forecasts, transforming the way of identifying and servicing high growth corporations.