Corona Daily Comment – ENG | 09

EXCLUSIVE! The results of the model developed by FinLantern on the progress of the Coronavirus epidemic

Coronavirus Daily Comment (issue #9)

(by Riccardo Esposito, CEO at FinLantern)






In this number:

– How the number of swabs affects the statistics of the infected
– Comment on yesterday’s data (March 29)
– Forecast on today’s data (30 March)
– Analysis of the possible impact of the latest data on the Italian market: NEUTRAL!
– (in the next issue) Update on the situation in Germany, France, Spain, UK, Switzerland and the USA


Dear FinLantern friends,

I start immediately by telling you that the data relating to the number of new infected in Italy confirm, based on the model we developed, that the peak should already have been reached at the end of last week and that now, maintaining the restrictive measures and compliance for the quarantine, it will be possible to arrive in Italy within 2-3 weeks to have some control of the epidemic with a drastic drop in new infections.

Our OPERATIONAL ADVISE ON THE FINANCIAL MARKETS therefore remains that of an overweight on Italy, while, in light of what was highlighted in our analysis in the previous Corona Daily (n. 8), we suggest underweighting for the events related to COVID -19 (excluding other factors) Spain and the USA.


Some remarks must be made, however, before being too much optimistic:

  • At the current levels of spread of the epidemic, it is still absolutely necessary to continue working on the reduction of the Ro index (which must be steady <0) in order to give more breathing space to hospital structures now collapsing (we have already written in our previous comments that, based on our calculations, there is a gap of about 500-750 beds for seriously ill patients for COVID-19 every day, therefore excluding all seriously ill patients for any other reason).
  • The reduction of Ro can only take place with the respect of the quarantine by everyone.
  • This reduction trend does not currently take into account the main potential risk associated with the emergence of new outbreaks, particularly in the South.
  • By “epidemic control” I mean a level of cases reduced to a few dozen a day (COVID-19 cannot in fact be eliminated entirely without the attainment of herd immunity or the creation of a vaccine) which can be immediately and promptly identified and isolated, thus reconstructing the chain of contagion (active and passive).
  • Our forecast, however, of a “return to normal” with the reopening of the activities could be optimistically expected by the end of April, but only if we will be able to implement the measures of rapid identification of positive cases, resorting not only to more advanced and rapid tests, but also to technology (App? GPS?). Furthermore, it will certainly help the fact of being able to find more effective treatments and to implement isolation and protection measures for the weaker and virus-prone age groups.



Finally, I am able to answer a question that has been asked by many of you: how is the number of swabs that is carried out daily considered in the FinLantern model in order to predict the evolution of the epidemic?

Effectively, the number of tests that are carried out has a direct correlation with the number of positive cases detected. Besides, if you don’t look for something, the probability of finding it will be much less … and viceversa.

This is one of the main reasons, as I explained in n. 1 of this Corona Daily, for which I had selected Italy as a school case for all western countries: the fact that it did a large number of tests from the beginning.

In the case of the model I developed, although the number of swabs does not fall among the variables taken directly into consideration, it should be stressed that the consistency of the results provided by our model remains intact for the following reasons:

  1. The model is based on the daily averages of the number of new infected. In this way, the effects of deviations that may occur from having made an abnormally high or low number of tests in a single day are amortized and diluted.
  2. The goal of the model is not to predict the punctual number of daily infections, for which there could be significant differences without affecting the goodness of the model itself, but to predict current trends. In this sense, it is important for the model that the test policies are not substantially changed over time.
  3. The forecast results of the model must however be able to be interpreted. In this important interpretation work on my part, the number of swabs made was always taken into consideration. To make you better understand the work I have done, in the chart below, the absolute number of daily swabs that have been carried out in Italy (left scale) has been related to the percentage of positive cases that have emerged thanks to them (right scale).


Confirming the good quality of the contagion data in Italy in the last few days highlighted by the trend of the Ro index (see our analysis in the previous reports), although these had remained stable around the maximum levels of 6,000 cases per day, it should be noted that also the qualitative verification of these numbers, considering the number of tests carried out, gives a positive indication: the contagion numbers of the last days in fact reveal that they have been obtained in the presence of a marked increase in the test activity (in the last 3 days about 50% more tests were done than the previous week). Furthermore, despite this higher and more capillary virus search capacity, it does not seem that there is a proportional growth of the new infections, as highlighted by the fact that their incidence has gone from about 24% to about 17%.

Tomorrow, I will give you an update on how the situation is evolving in Germany, France, Spain, UK, Switzerland and the USA.



COMMENT TO THE DATA of 29/03/2020


Yesterday’s data (29/03/2020) confirm that in Italy the exponential growth phase (Ro> 1) has been abandoned in favor of the linear one (Ro = 1), as we previously estimated. I believe that the data of the weekend, in which Ro seems to have stabilized around this value, must have a substantially NEUTRAL impact on the Italian market.

IMPORTANT! The target of a Ro of 1 was only an intermediate target. In order to be able to contain the epidemic, Ro must be significantly less than 1 and therefore resume its downward trend as soon as possible. I believe that the restrictive measures launched by the Italian government a week ago should begin to have an effect on the Ro precisely this week.


TODAY’s ESTIMATES (30/03/2020)

Our expectation for today is a total number of infected of around 102,000.



Here, in Switzerland today begins the third week of reclusion at home. I am spending a lot of time with my teenage daughter, trying to motivate her to take advantage of this situation anyway to broaden her horizons and develop new interests. But … ATTENTION! I tell you: if the schools are closed for a long time, it will be me and the other parents of teenagers to find a vaccine against COVID-19 before the scientists! 🙂

Keep on writing to me and asking me questions, and if you think my analyzes and comments are useful, don’t hesitate to share them with your friends and colleagues. For this purpose, I also created a sharing button on LINKEDIN:


A good #IStayHome to all,







Riccardo Esposito

CEO – FinLantern





NOTE! In order to receive the Corona Daily by email it is necessary to register on Otherwise you can always download it for free on